Navigation: Jump to content areas:


Pro Quality. Fan Perspective.
Login-facebook
Around SBN: This Week In GIFs

The Rainman Projections For The Franchise And Friends

NameWLERAGS GSVIPHERHRSOBBWHIPK/9BB/9K/BBWAREdit
Tim Lincecum 19 7 2.75 33 33 0 225.0 172 69 16 262 68 1.07 10.48 2.72 3.85 7.4 Edit
Matt Cain 16 10 3.25 33 33 0 225.0 202 81 22 188 68 1.20 7.52 2.72 2.76 4.6 Edit
Jonathan Sanchez 13 10 3.75 32 32 0 195.0 180 81 19 206 91 1.39 9.51 4.20 2.26 3.8 Edit
Barry Zito 13 10 4.25 32 32 0 195.0 191 92 24 141 80 1.39 6.51 3.69 1.76 2.0 Edit
Other Starters 9 15 4.75 32 32 0 165.0 172 87 21 119 68 1.45 6.49 3.71 1.75 1.6 Edit
Brian Wilson 4 1 2.75 0 75
41 75.0 58 23 5 88 31 1.19 10.56 3.72 2.84 1.9 Edit
Sergio Romo 4 4 3.25 0 75 0 75.0 68 27 5 79 18 1.15 9.48 2.16 4.39 1.9 Edit
Jeremy Affeldt 4 4 3.25 0 75 0 75.0 68 27 5 62 31 1.32 7.44 3.72 2.00 0.9 Edit
Brandon Medders 4 4 3.75 0 75 0 75.0 73 31 5 62 35 1.44 7.44 4.20 1.77 0.5 Edit
Other Relievers 4 7 5.25 0 153 0 153.0 152 88 17 112 88 1.57 6.59 5.18 1.27 -0.6 Edit
Total 90
72
3.74
162
162
41
1458.0
1336
606
139
1319
578
1.31
8.14
3.57
2.28
24.0 Edit
2009 88
 74 3.55
162
162
41
1446.0
1268
571
140
1302
584
1.28
8.10
3.63
2.23
21.7 Edit

This FanPost is reader-generated, and it does not necessarily reflect the views of McCovey Chronicles. If the author uses filler to achieve the minimum word requirement, a moderator may edit the FanPost for his or her own amusement.

Comment 227 comments  |  0 recs  | 

Do you like this story?

Comments

Display:

90 wins counting Spring Training?

Barry Zito: Mike Hampton with a guitar

by JakeS on Jan 31, 2010 9:28 PM PST reply actions  

Ah, that would explain how the win total looked high but the peripherals fairly reasonable. Yes, maybe we will win 90 games if Aubrey Huff, Edgar Renteria, and Mark De Rosa all bounce back together from lousy seasons despite their ages and our “other infielders” post an OBP of .314, better than our team OBP last year and Freddy Sanchez has a good a season as he did in his prime, and Bengie Molina has a .300 OBP and Pablo Sandoval is really a .330 hitter, period. None of those projections are necessarily unreasonable by themselves, but collectively, well that’s a lot of Kool-Aid.

Barry Zito: Mike Hampton with a guitar

by JakeS on Feb 1, 2010 9:04 AM PST up reply actions  

None of those projections are necessarily unreasonable by themselves

And to be fair to GRM, based on a discussion we had, it seems that’s kind of the point. These projections are what he thinks are the “most likely” level of performance from the players, but don’t take into account the probability of injury, regression, or players just falling off the cliff altogether. So each one might be pretty likely, but a whole team is going to experience some of the injury/decline stuff, so as a whole they’re probably pretty unlikely.

by Missing Barry on Feb 1, 2010 9:47 AM PST up reply actions  

so it would be completely

WRONG to look at these projections, add up the WAR and say “yup, 90 win team”.

And of course, it would be ridiculous to compare GRM projections to, say, any other projections, because they are not even trying to do the same thing.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 1, 2010 1:37 PM PST up reply actions  

I hope other projections systems aren’t trying to do the same thing. They should be accounting for probability of injuries, of breakout years, of down years, of players ability to play at an MLB level disappearing entirely, of deadarm, etc, and if they aren’t, then their team level wins projections are questionable (though not as much as you’d think because every team faces those risks, though the problem is with the varying degrees they face those risks).

by Missing Barry on Feb 1, 2010 1:54 PM PST up reply actions  

No, It Would Not.

For the most part teams (other then the $120M+ Payroll Teams) that make the playoffs do so not just because they have the needed talent but because they stay healthy and avoid key players having bad years. This is even true sometimes for these teams with $120M+ payrolls. This is the only way they Giants are going to make the playoffs when they do so. The Giants are never going to be so talented that this will not be needed as their budget can not afford this much talent.

by giantsrainman on Feb 1, 2010 2:42 PM PST up reply actions  

The Giants are never going to be so talented that this will not be needed as their budget can not afford this much talent

Disagree. We have more resources than anyone but the Dodgers in the NL West. We don’t have the Yankees or Red Sox in the NL taking away a wild card spot. The big money NL teams (Cubs, Mets) are poorly run. We can make the playoffs without a “lucky” amount of health/performance from our collective team. Not saying we should expect to overcome those “unlucky” years, or that we’ll make it every year we aren’t “unlucky”, but we definitely should make the playoffs with neutral or slightly bad luck at a fairly high probability.

Also, I think the problem with your projections is if you go through them individually, almost all the players have the possibility of going down (whether it be injury or a down year or their career imploding, as many of our players have a chance of happening), and by a significant amount, whereas none of them have the possibility of going that much higher. I think you’re looking at a very skewed distribution where the down/up probabilities aren’t even close to equal, and it leans heavily “down”.

For instance, you have Matt Cain at 225 innings for a 4.6 WAR. How much higher than 4.6 WAR can he possibly get? He can’t throw much more than 225 innings. His FIP will already be a career best to get that 4.6 WAR, is there much room for it to improve? Probably not. However, he could get injured and throw as few as (knock on wood) 0 innings (or anywhere in between). He could get “deadarm” and not perform as well. I’m pretty confident his peripherals have more room/probability to get worse than better. So on the whole, it’s unlikely he posts much more than 4.6 WAR, but he could post anywhere less all the way to 0. Those kinds of probabilities exist with nearly all our players – much higher downside (probability*loss of production) than upside (probability*increase in production), and at the team level, it’s very likely we’ll get hit with some of that “net downside”, even without bad luck/worse than expected injuries….

by Missing Barry on Feb 1, 2010 2:54 PM PST up reply actions  

You Ignore The Upside That Is There

Lincecum could top 10 WAR, Panda could top 7 WAR, and J.Sanchez could top 5 WAR. Five of the seven veteran hitters could return to the 4 WAR level these five have all reached in the last 3 years. And this upside does not even include Bumgarner and Posey and to a lessor extent the other non veterans.

by giantsrainman on Feb 1, 2010 3:19 PM PST up reply actions  

That seems unlikely.

No pitch has topped 10 WAR (at least not in the period fangraphs provides it for, I suspect Randy Johnson accomplished in once or twice between 1999 and 2002, but I can’t see Lincecum throwing the 250 or so innings he’d probably need to do it). Sandoval as a 7-WAR hitter is more likely, but I don’t think it’s more likely than him getting hurt and missing 1/3 of the season.

VAE PVTO DEVS FIO

by Bhaakon on Feb 1, 2010 4:19 PM PST up reply actions  

Thru The All-Star Break Timmy Was On Pace For 10+ WAR Last Year

I do not at all find it hard to believe that he could do this for a full season. In fact I expect that his peak year will be such a year.

by giantsrainman on Feb 1, 2010 4:32 PM PST up reply actions  

Which is better evidence for not considering a half season an acceptable sample size than it is for projecting a 10 WAR season from Lincecum.

And where do you get this 1st half projection (also recall that WAR is a counting stat, and the ASG tends to occur past the halfway point in the season, you can’t just take the WAR total at the ASG and multiply it by 2).

VAE PVTO DEVS FIO

by Bhaakon on Feb 1, 2010 4:41 PM PST up reply actions  

He Was Closer To 7 WAR Then To 6 Already

I don’t think SSS has anything to do with it. it was fatigue driven as shown by some drop in velocity. As Timmy continues to get stronger he will stop running out of gas and stay strong all year and thus have a peak year above 10 WAR. He is just that damn good!

by giantsrainman on Feb 1, 2010 4:53 PM PST up reply actions  

You sound an awful lot like Brian Sabean. “I’m going to find a split I like and project that as his true talent, completely ignoring the larger, better sample I have to look at”…

by Missing Barry on Feb 1, 2010 8:22 PM PST up reply actions  

Not the Same Arguement At all

It is frankly silly to think Timmy can’t get any better. The greats almost all do and he is very much showing signs that he is one of them.

by giantsrainman on Feb 1, 2010 9:12 PM PST up reply actions  

Everyone has a ceiling. At some point you simply cannot strike out any more batters than you already do, or walk fewer, or allow fewer HR’s. When you consider the lack of control pitchers have over BABIP (not total, but a lot), that’s a lot that he simply cannot improve upon. It’s silly to think guys have a high probability of improving on 8.2 WAR seasons. The fact of the matter is they’re more likely to go down from that then up. He pitched 225 innings last year, that can go down, but how much farther up can it go? There are a lot of things that contribute to creating a ceiling for players.

And yes, it is the same argument. You used first half splits, you ignored what he did in the second half, and found a reason to dismiss it (fatigue driven), then made a definitive statement that he’ll get over that problem. That is EXACTLY what Sabean does, and EXACTLY why we get pieces of shit like Edgar Renteria. He convinces himself of bullshit reasons to ignore better information in favor of worse information, as you just did. It’s complete ignorance of probability theory, and frankly, it’s assuming you know more than you possibly can. Sometimes it’s best to understand the limits of your knowledge. There are never definites in projecting the future.

by Missing Barry on Feb 2, 2010 8:07 AM PST up reply actions   1 recs

I agree with everything but...

the reason why we got POS Renteria. The reason we got him was that we had no viable SS at any level above A-ball. The FA options were all pretty bad, and Renteria was likely best player available. Sabes did pay too much, and Rent did fall off the proverbial cliff, but sadly this was one of his more defensible moves

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 2, 2010 8:23 AM PST up reply actions  

But I’m not talking about the logic as much as the evaluation of what kind of a player Renteria was. The line we got was some BS about his second half, and how somehow his first half was meaningless. That kind of evaluation had an effect on our targetting Renteria specifically, and how much we offered him, and frankly, that kind of evaluation sucks. You can’t just ignore evidence that doesn’t fit your argument/evaluation.

by Missing Barry on Feb 2, 2010 9:12 AM PST up reply actions  

You Ignore The Upside That Is There

I don’t see where you get that from. I pretty clearly acknowledge the possibility for upside, but the whole point is the up isn’t as big or as likely as the down. There’s a reason for “regression towards the mean”, because players go towards it more often than they go away from it, it’s a simple fact of probability. Plus, injuries bring you down, there is no opposite of an injury that makes you a better player (ever since they banned the Cream and the Clear, am I right?!).

The whole point is the upside is there, but it’s not as big as the downside, nor is it as likely.

by Missing Barry on Feb 1, 2010 8:25 PM PST up reply actions  

You Don't Even Understand Where The Mean Is For Each Player

To treat them all as if they all have the same mean is silly.

by giantsrainman on Feb 1, 2010 9:13 PM PST up reply actions  

I initially typed “regression to the mean”. I then consciously changed it to “regression towards the man” as an acknowledgement of this very concept. The thing is, I understand that it’s impossible to know where each individual players mean is. I have strong doubts that you understand that concept. All we have are sample sizes of data to try to estimate that – this is basic statistics at its core – and the fact is, when you’re talking about performances not at the overall mean, your sample comes from a part of the distribution that’s skewed – that is, there are less observations further from the overall mean than closer to the mean. That’s why we “regress towards the mean”. It’s not because we assume every player is at the overall mean, that would be silly, there are clearly different levels of talent – it’s because we understand we’re working with samples and we don’t have an even distribution for the individual we’re looking at – it’s skewed, and it’s likely that his individual mean is closer to the overall mean than farther. It’s an acknolwedgement that we can never know a players true mean, so we play the odds the best we can. That’s how you maximize decision making and performance. And that’s the concept you fail to understand entirely.

by Missing Barry on Feb 2, 2010 8:12 AM PST up reply actions  

regression towards the mean

by Missing Barry on Feb 2, 2010 8:13 AM PST up reply actions  

Any idiot can regress toward the mean. Only true superstars can regress toward the man.

by Evan on Feb 2, 2010 8:27 AM PST up reply actions  

Yeah maybe regression toward the man IS appropriate in Lincecum’s case…

by Missing Barry on Feb 2, 2010 9:13 AM PST up reply actions  

My old lady

is good, but no superstar

by JetSam on Feb 6, 2010 5:14 PM PST up reply actions  

but so what?

Any 82 win (true talent level) team can have a good year and make the playoffs. For example, the 2009 Giants. Well, they were a couple of games out, but in some divisions/years 88 wins is enough.

So what, exactly, is the difference in some computer projection system which crunches the numbers and gets a 83 win season and you going off on the mountian, coming back and saying “if everything goes right they could win 88-90 games”. Duh. That’s what a projection of 82 wins-on-average means.

The difference is that the other projection is more realistic, while yours is just pie-in-the-sky.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 1, 2010 4:09 PM PST up reply actions  

You’re projecting the pitching to be better than it was in 2009? Our top 4 throwing 840 innings would be nice though.

Juan Carlos Perez, please start hitting.

by marcello on Jan 31, 2010 9:33 PM PST reply actions  

You’re projecting the pitching to be better than it was in 2009?

Not talking about ERA or WHIP.

Juan Carlos Perez, please start hitting.

by marcello on Jan 31, 2010 9:40 PM PST up reply actions  

Yeah, I Am Projecting A Small Improvement In WAR Driven By FIP

SOs go up form 1302 to 1319, BBs go down from 584 to 578, and HRs go down from 140 to 139.

by giantsrainman on Jan 31, 2010 9:44 PM PST up reply actions  

Also innings go up from 1446 to 1458.

by giantsrainman on Jan 31, 2010 9:45 PM PST up reply actions  

Less Defense Is The Reason I See ERA And WHIP Going Up

Last year we had a +51 UZR Defense and I think this year it will be just +24.

by giantsrainman on Jan 31, 2010 9:55 PM PST up reply actions  

Wait, what?
You Are Way Over Analyzing This

Thus far we have replaced two players on our position player roster with two newcomers. We have replaced Randy Winn in the OF with Mark DeRosa and we have replaced Ryan Garko in the IF with Aubrey Huff. Career stats for both of changes show that the new guys are basicly the same defensively as the old guys. Bottom line, there is zero reason to expect any significant change in the performance of the Giants defense.

I was promised lasagna.

by Cookyman on Feb 1, 2010 12:44 AM PST up reply actions  

Yeah, I Realized I Was Wrong About That. Rare I Know For Me To Be Wrong But There You Go. :-)

Some might say even rarer for me to admit being wrong though I really can’t see how anyone could justify that opinion. :-)

by giantsrainman on Feb 1, 2010 12:49 AM PST up reply actions  

Ah, OK

I was promised lasagna.

by Cookyman on Feb 1, 2010 9:14 AM PST up reply actions  

Thanks. It’s always interesting to see other people’s projections. You’re a lot more optimistic about the pitching than I am.

I try not to break the rules, but merely to test their elasticity.

by esseffgeez on Jan 31, 2010 10:26 PM PST reply actions  

That’s a spiffy looking chart

Matt Cain: He'll save children, but not the Dodger children.
"Out, out, Fred Lewis!" - JCTillam Gamerspeare

by jponry on Jan 31, 2010 10:33 PM PST reply actions  

It seems fairly reasonable, though I think the distribution might be a bit different in terms of WAR

Matt Cain: He'll save children, but not the Dodger children.
"Out, out, Fred Lewis!" - JCTillam Gamerspeare

by jponry on Feb 1, 2010 12:20 AM PST up reply actions  

I Really Can't Take Credit (Or Blame) For The WAR Estimates

Fangraphs “Fans Projection Tool” calculated these. What i provided were my projections for role, innings, K rate, BB rate, and HR rate.

by giantsrainman on Feb 1, 2010 12:28 AM PST up reply actions  

This I can agree with.

Barry Zito: Mike Hampton with a guitar

by JakeS on Feb 1, 2010 9:07 AM PST up reply actions  

Lincecum will be better than that, Cain will be a little worse than that, Sanchez will be a little worse than that, Zito will be a little better than that, Starter #5 will be a little better than that.

I’d expect a lot fewer starter’s wins and a lot more reliever’s wins given the quality of our offense. And I doubt the rest of the ‘pen is worth -0.6 WAR; we’ve got a lot of quality young arms coming up, especially Runzler and Joaquin.

by quincy0191 on Jan 31, 2010 11:01 PM PST reply actions  

What Is Your View On How Many Runs We Will Allow In 2010?

We allowed just 611 in 2009 and I am forecasting this to increase to 648 (606 earned + 42 unearned) in 2010 driven mostly by less defense (Team UZR dropping from +51 to +24).

by giantsrainman on Jan 31, 2010 11:13 PM PST up reply actions  

you projection is optimistic

both in IP and RA.

You are not acounting for the fact that there is a probability of something bad happening to drag the numbers down.
For example, you have Tim and Matt at 225 IP. This is clearly something like a maximum. There is a much greater chance of them pitching only 200 or 180 or even 150 than 240+.

This is true of your hitters as well… it’s a “if everything goes right”. To be fair, you might have a higher chance of predicting a few of these guys exactly correct (if we generously assume that you are predicting the “mode” or most probable outcome for each player)…. but those kind of projections are totally useless for team performance, because you are GUARENTEED that some people are going to do worse (and some better), but since your projections are on the high end it’s much more likely for someone to fall short of your projections than exceed them.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 1, 2010 9:19 AM PST up reply actions  

My favorite part

These projections, which are obviously very optimistic, still only get us to 90 wins.

Juan Carlos Perez, please start hitting.

by marcello on Feb 1, 2010 9:54 AM PST up reply actions  

yeah

I think if you did the same for the dodgers they would probably project to 108 wins.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 1, 2010 1:37 PM PST up reply actions  

2009 Forms A Good Arguement That These Pitchers Projections Might Be Pessimistic.

I can understand someone concluding that my hitting projections in total are optimistic (though I disagree) but I think you just have to be a negative nilly to do so with regards to these pitching projections in total. Can one take issue with some of the individual pitcher’s projections? Sure, but in aggregate I don’t see much of a basis other then pure pessimism to do so.

by giantsrainman on Feb 1, 2010 2:33 PM PST up reply actions  

You have them projected to be 2 wins better than last year. That’s pretty damn OPTimistic.

Juan Carlos Perez, please start hitting.

by marcello on Feb 1, 2010 3:00 PM PST up reply actions  

Ignore The 24 WAR As I Do Not Buy Fangraphs Calculation Here

I have them pitching 12 more Innings with 17 more Strikeouts, 6 fewer Walks, and 1 Fewer HR. Basicly I have them being the same overall as a team. I just do not buy that these four small changes add up to 2.3 WAR.

by giantsrainman on Feb 1, 2010 3:10 PM PST up reply actions  

it’s actually the opposite. The individual projections look OK, on the surface.
But every single one is projected to have an ERA on the good side and an IP on the high side.

Compare Zips:
Linc: 2.65, 225 (actually the same)
Cain 3.51 217 (not too far, but a 1/4 run worse than yours)
Sanchez 4.15 156 (way off)
Zito 4.48 184 (significantly worse than you)
Wilson 3.47 70 (Wilson only has 1 good season under his belt)
Romo 3.21 53 (ERA close, but many fewer innings)
Affeldt 3.48 64 (another 1/4 run here)
Medders 4.65 56 (a whole run worse).

So first off – you have projected the top 4 relievers to pitch 300 innings. That’s unrealistic. Last year, the Giants top 4 releivers in IP had only 260 or so. And they had one of the top relief staffs in the league.

Second, you project Sanchez to pitch 180 innings, even though he’s never done it. Ever. Now can he? Yeah, sure. But I sure as heck wouldn’t PROJECT him to bump up his IP 20%.

So you have at least 100-150 IP here allocated to really good pitchers, when it’s quite likely that they will get distributed to worse parts of the staff. Plus the ERAs are uniformly optimistic.

So that’s where you see the fact that your individual ERA projections are OK (if you assume Medders is FIP immune, and Zito is back on the horse)

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 1, 2010 4:29 PM PST up reply actions  

ZIPs Is Also Way To Slow to Recognize Breakout

They didn’t see Panda’s last year and still don’t fully see it this year. Likewise they have failed to see the “real” breakout Sanchez’s and Wilson’s seasons last year represent.

As i have stated I can justify expecting some “regression” but really just adjustment to the new reality of less defense resulting in slight increases in team ERA and WHIP but i can not see any justification for projecting increases of any significance in team FIP.

by giantsrainman on Feb 1, 2010 4:43 PM PST up reply actions  

but you don't KNOW they are breakouts

you won’t know until their careers are basically over.

You are just assuming they are breakouts. Also known as wishcasting. Also, Sandoval’s projections are pretty good.

Basically what you are saying is that "ASSUMING no one gets hurt and ASSUMING Panda, Sanchez, and Wilson’s breakouts are real and permanent and ASSUMING Zito has reattained a new level AND a host of other assumptions that all favor the Giants players performing on the high side of their potential (I won’t say optimal… they are not that pie-in-the-sky) then they will win 90 games or so.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 2, 2010 8:28 AM PST up reply actions  

You have no evidence that this is true other than the fact that you simply feel that Sandoval “should” have a better projection.

In MLB history, there are 297 instances (at the major league level), of:

- Player has 3 consecutive seasons of 300 at-bats or more.
- Player improves by 100 points of OPS in year 2 from year 1
- Player is 25 or under during the improvement year.

The average improvement of these players was 152 points of OPS. On average, in the year after the improvement, these under-25 players maintained roughly 2/3rds of their gain in their following seasons (65.8%).

Sandoval’s combined major league performance and translations in 2007 and 2008:

2007: .680 OPS
2008: .797 OPS

ZiPS projected Sandoval in 2009 to have a .768 OPS, retaining 70.1% of his gain.

2008: .797 OPS
2009: .943 OPS

ZiPS projects Sandoval in 2010 to have an .884 OPS, retaining 71.9% of his gain.

In other words, if you feel that Sandoval is going to outperform his 2010 ZiPS projection, that’s certainly your prerogative.

But, if you’re going to argue that ZiPS has a specific systemic problem with evaluating players that have breakouts, you have to back it up or the null assumption is that you’re being intellectually dishonest.

--
Dan Szymborski
dan@baseballprimer.com

by D.Szymborski on Feb 2, 2010 8:50 AM PST up reply actions  

GRM has a hard time with probability distributions. They don’t seem to exist in his world.

by Missing Barry on Feb 2, 2010 9:14 AM PST up reply actions  

I should also note that, of those 297 players that gained at least 100 points of OPS in the second season, 291 of 297 gained less than the 143 additional points of OPS Sandoval gained in the third season and only a single player did it in recent history (J.D. Drew from 1999-2001).

Sandoval’s improvement has been of the magnitude that is exceedingly rare. Projection systems shouldn’t be projecting exceedingly rare events as the most likely ones.

--
Dan Szymborski
dan@baseballprimer.com

by D.Szymborski on Feb 2, 2010 10:29 AM PST up reply actions  

Actualy I Do Have Evidence - The Evidence Of My Eyes And More Importantly Professional Eyes

Stats are too slow to see this. I don’t blame them for this i just Recognize it is true.

by giantsrainman on Feb 2, 2010 4:34 PM PST up reply actions  

The thing is, you don’t realize that those are subject to probability distributions as well….

by Missing Barry on Feb 2, 2010 5:06 PM PST up reply actions  

Which Require Way Too Large A Sample Size To Be Useful

Decisions have to made in real time with the sample size available and in my view these decisions are best made by human talent evaluation not by statistical formulas.

by giantsrainman on Feb 2, 2010 5:12 PM PST up reply actions  

As we clearly saw when the statistics revolution took off to replace “human talent evaluation” and failed miserably because they already had it down perfect! It’s really too bad, I heard that Michael Lewis book could have been good if he finished it.

by Missing Barry on Feb 2, 2010 6:40 PM PST up reply actions  

Apples And Oranges

One takes about how to measure more accurately the past and determine what is and is not valuable the other talks about how to predict more accurately the future.

by giantsrainman on Feb 2, 2010 6:58 PM PST up reply actions  

Looking at the past to determine value isn’t the only point to statistics, though. They are useful for predicting the future. There’s a reason many of the most successful clubs are statistically orientated, and use statistical techniques for predicting the future. There’s a lot of value in it. When you have players at the MLB level with large samples of production – there’s not a whole lot there to be scouted. Combine that with human’s inability to objectively form impressions of anything without systematically and objectively recording it (which turns it into statistics), and I just really don’t know why you think there’s some infiinite ability in scouting to determine the future. What’s watching a guy at the plate really going to tell you? Maybe you see they’ve lost some bat speed, fine, but how much, exactly, and what exact effect is that going to have on the future? It’s still a guessing game trying to play probabilities, and you’re still just looking at skills without a perfect idea of how that translates to “production”, and a lot of skills like plate discipline just really don’t change for most players over the course of their career. What does that tell you about what they’ll do in the future?

It’s not that I’m discounting scouting, I think it’s very, very useful for players without an MLB track record, I just don’t see what makes you think it can perfectly predict players talent levels (or not even perfectly, but just see changes in talent levels accurately at all) who have an established MLB track record? More information is better, scouting reports should be used, but they’re hardly an end-all for looking at these guys.

by Missing Barry on Feb 3, 2010 7:18 AM PST up reply actions  

Do you not remember that the same Scouting Eyes saw a breakout from Velez last year, when the stats said he’s the same? Or when they saw Durham collapse in 2007, and the stats indicated he’ll be OK the following year? What about in 2008, when they saw Swisher collapse, but the stat crowd said he’ll be good as new the following year? Why does this mean nothing to you?

I was promised lasagna.

by Cookyman on Feb 3, 2010 10:32 AM PST up reply actions  

I’m not even sure even stated anything, let alone overstated. You’re claiming that stats are only useful for looking at the past, not for predicting the future. I brought up three examples from the past 2 years of stats predicting the future. I can bring you 10 more if you want.

I was promised lasagna.

by Cookyman on Feb 4, 2010 3:57 AM PST up reply actions  

and that is why

any prediction for team wins from your player projects are too high

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 2, 2010 8:25 AM PST up reply actions  

no

the good things and bad things have to cancel.

Because the ONE thing I am sure of is that some guys will do better than their (average) projections and some will do worse.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 2, 2010 4:58 PM PST up reply actions  

Not The Way I See It

There is upside above my projections that does provide balance.

by giantsrainman on Feb 2, 2010 5:09 PM PST up reply actions  

I expect my boss to give me a $10 million bonus when I show up to work tomorrow. But, the good news is, there is upside there. He might give me $20 million.

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 5:34 PM PST up reply actions  

You have yet to acknowledge the upside is less probable, and if it happens it will be to a lesser degree, than the downside…

by Missing Barry on Feb 2, 2010 6:41 PM PST up reply actions  

I’ve gone over why it’s true – from injuries to ceilings, from the difference between 0 and 225 innings compared to 225 and a maximum of what, 235?

by Missing Barry on Feb 3, 2010 7:19 AM PST up reply actions  

But some things WILL go wrong – that’s a given. Somebody’s going to get hurt. Somebody’s going to slump. Something’s going to go wrong.

My Bucardo is better than yours.

A hot August weekday, before a small crowd, when the only thing at stake is the tissue-thin difference between a thing done well and a thing done ill. Insofar as the clutch hitter is not a sportswriter's myth, it is a vulgarity, like a writer who writes only for money.

by Roger on Feb 2, 2010 5:05 PM PST up reply actions  

On Offense To An Extend With Regards To Pitching Not At All

However even on offense I argue that my projections (collectively not necessarily individually) are probable if the key eight players I identify have no health issues.

by giantsrainman on Feb 2, 2010 5:28 PM PST up reply actions  

Which would make them not projections at all.

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 5:35 PM PST up reply actions  

Nope - Just Means Not Limited By A Matmematical Formula

Thus free to consider things not accounted for in said mathematical formula.

by giantsrainman on Feb 2, 2010 6:03 PM PST up reply actions  

You know you can add “things not accounted for in said mathematical formula” INTO a mathematical formula, right? It’s really not that difficult, you just add in the thing not accounted for as a new variable…

by Missing Barry on Feb 2, 2010 6:42 PM PST up reply actions  

PECOTA does

Using signing bonus as a proxy. It’s not a flawless way of doing it, which means you’ll dismiss it, but short of somehow getting scouting reports on every single player, it’s pretty good.

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 7:49 PM PST up reply actions  

Way Too Limited

Where is the input on what the scouts have to say about what he has done since or even more importantly recently?

by giantsrainman on Feb 2, 2010 7:51 PM PST up reply actions  

I forget which projection system did it, but one of the guys doing projections last year went through his list and looked at players that just “looked off”. He noted them, added a binary variable to everyone, gave those guys he thought were off a 1 and everyone else a 0 (so their projection didn’t change)….and it turned out his adjusted projections (he was just doing it as a test to see how it did, but discussed it in a blog somewhere) did better than the other ones. So like I said, you can include that kind of thing in an equation.

I’d also like to know what, exactly, you think scouting tells us? Like if you had a projection system, and scouting reports, what type of scouting report variables would you include in the projection system?

by Missing Barry on Feb 3, 2010 7:22 AM PST up reply actions  

Really?

Let’s have a quick tour around 2009 to check and see:

LA Dodgers: Manny Ramirez suspended for 50 games, Chad Billingsly hurt and terrible much of the end of the season

Colorado: Jeff Francis out for year, Garret Atkins, horrible

Cards: horrible years from opening day starters Ankiel, Greene, Duncan, and Barden.

Phils, terrible seasons from Rollins, Hamels, and Lidge

Angels, massive injuries to starting rotation

Twins, Justin Morneau misses last two months of the season

That’s hardly an exhaustive list — just things I could come up with off the top of my head.

My Bucardo is better than yours.

A hot August weekday, before a small crowd, when the only thing at stake is the tissue-thin difference between a thing done well and a thing done ill. Insofar as the clutch hitter is not a sportswriter's myth, it is a vulgarity, like a writer who writes only for money.

by Roger on Feb 2, 2010 6:53 PM PST up reply actions  

They who? The Twins? The Rockies? I just listed 6 playoff teams from last year — every one that fit your criteria of being under $120 million (which is in itself arbitrary). I could have added the problems of the Yankees and Red Sox to make it a complete sweep.

I could also probably go back to 2008 or 2006 or 2002 and do the same.

How is a complete sweep of teams fitting your criteria counting as an “exception” precisely?

My Bucardo is better than yours.

A hot August weekday, before a small crowd, when the only thing at stake is the tissue-thin difference between a thing done well and a thing done ill. Insofar as the clutch hitter is not a sportswriter's myth, it is a vulgarity, like a writer who writes only for money.

by Roger on Feb 2, 2010 7:17 PM PST up reply actions  

They Being The Dodgers

The Phillies payroll was well above $120M. You are over stating the significance of the other loses by the other teams.

by giantsrainman on Feb 2, 2010 7:56 PM PST up reply actions  

Wrong again

Dodgers

Phillies

So the Phillies were over, if you add in players not actually on their team, and the Dodgers weren’t. Of course, you’re still ignoring the other teams Roger listed.

Look, you got caught in something you made up, once again. For once, act like an adult and admit you were wrong. Then learn to stop making shit up. This is fucking infuriating.

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 8:36 PM PST up reply actions  

The Dodgers Were In Essense - Backloaded Contracts

Again the significance of the loses by the other teams are being overstated.

by giantsrainman on Feb 2, 2010 8:46 PM PST up reply actions  

Be More Percise On The Dodgers

Orlando Hudson’s earned bonuses of $4.62M not included
Randy Wolf’s earned bonuses of $3M not included
Rafael Furcal’s Deferred signing bonus of $3M not included
Casey Blake’s earned bonuses of $0.5M not included

This $11.12M added to the $109.18 that is included totals $120.3M

by giantsrainman on Feb 2, 2010 9:29 PM PST up reply actions  

The Angels Were Also In Essense A $120M Payroll Team

$116.71M Before bonuses.

Bobby Abreu’s earned bonuses of $1M not included

This puts the Angels just short of $118m which is in essense a $120M Payroll team.

by giantsrainman on Feb 2, 2010 9:45 PM PST up reply actions  

So Where Are We

In my view Phillies, Dodgers, and Angels were all $120M Payroll teams in 2009. In my view the impacts of the loses listed to the Rockies, Cards, and Twins were minimal to their playoff chances.

by giantsrainman on Feb 2, 2010 9:53 PM PST up reply actions  

I think we've all seen ....

that it’s been proven that GRM is full of shit. He makes up stuff and tells you that you are wrong because he sees it with his eyes. What a crock of shit!

Why isn't Sabean held accountable for leading the Giants into many years of mediocrity???

by oldrips on Feb 2, 2010 10:14 PM PST up reply actions  

There’s just no way you’re not being disingenuous, which I don’t get from somebody who obviously has the faculties you do. So I’ll stop after this, but:

Losing your former MVP cleanup hitter for the stretch run is not a minimal impact (most people assumed it was the Twins death knoll when it happened). Losing your #1 starter for the season is not a minimal impact. Having to virtually recreate your starting lineup on the fly in the middle of the season (as the Cards did) is not a minimal impact.

By no conceivable definitions are these things minimal. They are the equivalent of say, Matt Cain missing the season, Pablo Sandoval missing August and September, or DeRosa, Schierholtz, Renteria and Huff all having such terrible seasons that they need to be replaced in the lineup (ooh, maybe I shouldn’t jinx us like that. Pardon me while I go outside, turn three circles and spit on the ground).

My Bucardo is better than yours.

A hot August weekday, before a small crowd, when the only thing at stake is the tissue-thin difference between a thing done well and a thing done ill. Insofar as the clutch hitter is not a sportswriter's myth, it is a vulgarity, like a writer who writes only for money.

by Roger on Feb 3, 2010 6:26 AM PST up reply actions  

Also, (one more), I don’t want to give the impression that 2009 was particularly unusual so:

2008

Chicago White Sox: lose leading MVP candidate Carlos Quentin for month, constant injuries to Creded, off year from Swisher

Tampa Bay: lost Carl Crawford for 50 games, tremendous power drop off for BJ Upton

Brewers: Rickie Weeks misses last two months, Eric Gagne flops as closer, Bill Hall still terrible, Jeff Suppan terrible

2007 DBacks, Chad Tracy injured, huge setback seasons for Drew, Young and particularly Quentin

Cubs: Ramirez, Soriano, Jacque Jones and Cliff Floyd all miss significant time with injuries

Angels, Matthews and Hillenbrand define awful, Bartolo Colon and Joe Saunders miss half season

Indians: Cliff Lee and Josh Barfield flop all the way back to the minors. Both corner OFs (Trot Nixon and Jason Michaels) terrible

And on and on it goes. People slump. Injuries happen. The inevitable results of the long season that must be dealt with through depth, through career years, through organizational flexibility, through deadline deals.

My Bucardo is better than yours.

A hot August weekday, before a small crowd, when the only thing at stake is the tissue-thin difference between a thing done well and a thing done ill. Insofar as the clutch hitter is not a sportswriter's myth, it is a vulgarity, like a writer who writes only for money.

by Roger on Feb 3, 2010 6:56 AM PST up reply actions  

Not Being Dissingenuous - But You Do Deserve A More Complete Response

I believe I have shown above how The Phillies, Dodgers, and Angels are among the $120M payroll teams and thus had the financial where with all to overcome significant loses. Now I will provide some details behind my assertion that the loses the Twins, Rockies, and Cardinals experienced were not significant.

Rockies – Yes they went into the season without Jeff Francis who had indeed been their ace as recently as 2007. However they only had to pay Jeff $3.75M and thus had the dollars available to pay for Jason Marquis as a replacement. And a fine replacement he was providing the Rockies 3.8 WAR which is virtually the same as the 4.1 WAR Francis provided in 2007.

Twins – Yes they played the last month without Morneau and you actually forgot to mention the first month without Mauer. However these two still combined to produce 11.4 WAR which is more then they ever combined to produce before in any season. So please stop crying a river for the poor Twins.

Cardinals – Yes they had four opening day position player starters have bad seasons. However, guess what? So did the Giants and we too were able to stay in the race all year even though we did not have the luxury of playing in the weakest division in the NL and maybe even all of MLB like the Cardinals did.

Bottom line, none of the obstacles these three teams overcame were anywhere near as significant as you portrayed them to be.

by giantsrainman on Feb 3, 2010 7:17 AM PST up reply actions  

Bottom line, none of the obstacles these three teams overcame were anywhere near as significant as you portrayed them to be.

But I never did attach a significance to them. I merely pointed out that they happened, and will happen in a majority of cases.

Here was my original statement:

But some things WILL go wrong – that’s a given. Somebody’s going to get hurt. Somebody’s going to slump. Something’s going to go wrong.

You then did a quick wheel around of your argument to this:

For Non $120M+ Payroll Teams That Make The Playoffs This Is Rarely True

which as was quickly pointed out implied that your “projections” weren’t an attempt at accurate forecasting at all, but rather simply a blueprint of what had to go right in order for the Giants to go to the playoffs, despite your vociferous defense of your objective accuracy in other parts of this fanpost.

To top it off, you’ve gotten so tied up in your pretzel logic that you quite incisively argue against yourself in the last post by saying this:

Yes they had four opening day position player starters have bad seasons. However, guess what? So did the Giants and we too were able to stay in the race all year

Quite right, the Giants could very easily have gone to playoffs last year with a couple lucky breaks here or there, despite a lot of things going wrong.

Your point about the Twins is also pretty much in line with what I suggested above, since some things are going to go wrong, other things have to go really really right(in the Cards case it was both the trades and as you very rightly point out, the lackluster nature of the rest of the division — that’s a key thing that can right for teams to help them overcome obstacles, see the White Sox in ’08).

But because things do go right, it does not therefore mean that the things that went wrong weren’t significant. It just means they were overcome.

Nonetheless, I restate my first position: things will very likely go wrong. Do with it what you will.

My Bucardo is better than yours.

A hot August weekday, before a small crowd, when the only thing at stake is the tissue-thin difference between a thing done well and a thing done ill. Insofar as the clutch hitter is not a sportswriter's myth, it is a vulgarity, like a writer who writes only for money.

by Roger on Feb 3, 2010 7:54 AM PST up reply actions   1 recs

When you talk about the probability of something bad happening, how do you expect him to project that? If he says somethign like, “One of our starters is likely to miss three games because of injury”, is he supposed to use a dart board to decide who it will be?

"We're in this thing!" My adopted Giant: "Raptor Jesus" Guzman, "Sweet Jesus" Guzman and Jesus H. Guzman.

by Goofus on Feb 1, 2010 4:29 PM PST up reply actions  

no

he’s supposed to amortize the IP lost across the 1-4 starters. So if there is a 10% (making this up) of a pitcher missing 90% of the season, then you would dock all starters 9% of their innings.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 1, 2010 4:32 PM PST up reply actions  

For individual pitchers, maybe

But overall, it’s likely to be more accurate.

I was promised lasagna.

by Cookyman on Feb 1, 2010 4:47 PM PST up reply actions  

No, it is more accurate. If your probabilities are in the ballpark, your projection taking the injuries into account will be more accurate – it will be correct (or closer to correct) a higher percentage of the time. Saying it’s not is simply untrue and distorting the facts.

by Missing Barry on Feb 1, 2010 8:29 PM PST up reply actions  

You Have A Funny Definition Of Accurate

In my book being intentionally too low so that you are closer to being right when things go wrong is way too far away from accurate to be of any interest to me.

by giantsrainman on Feb 1, 2010 9:16 PM PST up reply actions  

If by funny definition you mean I’m right a higher percentage of the time, sure. On any given individual, whatever, it’s not too important, but as you start getting a bigger sample size (like maybe a 25 player sample size, for instance?), it starts to matter. By being more accurate (projecting reality as opposed to whatever you call what you do), you maximize your decision making, and your results will be better. So yeah, I hardly think running a team to maximize your returns is “a funny definition”.

by Missing Barry on Feb 2, 2010 8:15 AM PST up reply actions  

I would have thought that the recent issues with financial markets not properly assessing risk regarding securities would have served as a cold shower for “These are the projections assuming nothing goes wrong!” genre of projections.

--
Dan Szymborski
dan@baseballprimer.com

by D.Szymborski on Feb 2, 2010 10:35 AM PST up reply actions  

One slight OT observation that I'm sure you've heard before

It is difficult to read your comments When You Capitalize Every Single Word In The Sentence.

by snafu on Feb 1, 2010 2:25 AM PST reply actions  

I Agree !00%

I don't know about that, to the groin.

by howtheyscored on Feb 1, 2010 8:16 AM PST up reply actions  

Actually, I just wanted to make that joke. It doesn’t bother me.

I don't know about that, to the groin.

by howtheyscored on Feb 1, 2010 8:17 AM PST up reply actions  

Leave Him Alone

That’s his trademark.

"We're in this thing!" My adopted Giant: "Raptor Jesus" Guzman, "Sweet Jesus" Guzman and Jesus H. Guzman.

by Goofus on Feb 1, 2010 4:30 PM PST up reply actions  

You wouldn’t ask Goofus to raise his humor level above eighth grade, would you?

Bruce Bochy would like you to look at the career numbers and stop complaining.
Bob Howry's #1 (and only) fan!!!
The Merkin Valdez of McCovey Chronicles!!!!!!

by cheno on Feb 2, 2010 12:32 AM PST up reply actions  

What effect do you think possible injuries will have, as well as any guys potentially having a really down year (and conversely, a career year, though based on your projections and methdology we discussed in your first projections it seems like down is much more probable than up for almost all of our players)?

by Missing Barry on Feb 1, 2010 7:29 AM PST reply actions  

Well, the general idea I’ve always gone with is that you can’t really predict career years or really sudden implosions so you just hedge your bets with by not forecasting any. Obviously you’ll have a few either way, but most likely they’ll just cancel out so your middle-of-the-ground projection is the safest.

As you say though, when you have a team with as many guys on the wrong side of 30 though, it’s less likely to see one of the career years than one of the implosions.

Barry Zito: Mike Hampton with a guitar

by JakeS on Feb 1, 2010 9:09 AM PST up reply actions  

Well, you can’t predict it, but you can still factor it in. If there’s a 10% chance of a guy having a career year, and a 10% chance of him having a horrible year, it’ll cancel out and not change your projection at all. If the chances are higher one way or another (for most of our players, it’s probably down), then you factor the level of production * likelihood of it happening to get a forecast that takes it into account….

by Missing Barry on Feb 1, 2010 9:49 AM PST up reply actions  

My initial reactions are...

-It would be awesome if Sanchez can do that
-It would be just as awesome if we could stay that healthy

by Sgt. Dingleberry on Feb 1, 2010 7:43 AM PST reply actions  

I would argue that the 5th starter could change that line significantly. Pucetas and Martinez rely on putting the ball in play, which could be more problematic with the changes in defense. If Bumgarner is the 5th starter, assuming the velocity comes back, that could put less pressure on the defense.

Proud adoptive daddy for the Big Unit. Thanks for sharing a year of your great career with the Giants and their fans.

by Speedforthewin on Feb 1, 2010 8:34 AM PST reply actions  

Except Bumgarner’s 6.3 (5.8 in AA) K/9 last year suggests he relies on putting the ball in play, too (in the present). At least Martinez makes sure those balls in play are on the ground…

by Missing Barry on Feb 1, 2010 9:51 AM PST up reply actions  

Where they can be efficiently gobbled up by… Edgar Renteria?

Barry Zito: Mike Hampton with a guitar

by JakeS on Feb 1, 2010 10:26 AM PST up reply actions  

Nah, but if it’s on the ground, it can’t go for a HR! Or really even a double, for that matter…

by Missing Barry on Feb 1, 2010 10:28 AM PST up reply actions  

It’s not.

I was promised lasagna.

by Cookyman on Feb 1, 2010 4:25 PM PST up reply actions  

Giants pitchers in 2009

Ground balls (not including bunts): .220/.220/.238.
Fly balls (not including line drives): .199/.194/.542.

Not even close. Ground balls are always better than fly balls.

I was promised lasagna.

by Cookyman on Feb 1, 2010 4:29 PM PST up reply actions  

No, there isn’t.

First off all, trying to pitch a RHH high and outside (which is what you seem to be implying pitchers should do) does not necessarily result in a fly ball to RF. Hitters can hit that pitch to CF. And if you make a mistake, they can pull it, too. That’s what you should be worried about.

There is not split that’s as specific as what you’re demanding. The good thing is that it doesn’t matter. Every single Giants pitcher does better on ground balls than on fly balls

Matt Cain:
.216/.216/.240 on ground balls.
.192/.188/.495 on fly balls.

Tim Lincecum:
230/.230 /.258 on ground balls.
192/.188/.495 on fly balls.

Jonathan Sanchez:
.261/.261/.289 on ground balls.
.218/.214/.582 on fly balls.

Again, non of these is even close. RF or not, it doesn’t matter. Overall, it is still much much better to get ground balls. Even if you think that fly balls to RF are better than ground balls (which I find very very hard to believe – it would require something like a 400 point difference in SLG between fly balls to RF and to LF. That’s insane), overall it doesn’t matter. If you’re counting on Bumgarner to be better on fly balls than ground balls, you’re counting on him to achieve something no Giants pitcher has even come close to achieving, including the best pitcher in the NL, and a good pitcher who lives on fly balls.

I was promised lasagna.

by Cookyman on Feb 2, 2010 2:42 AM PST up reply actions  

Your Numbers Are Distorted.

Homeruns as well as strike outs and walks should be excluded for this data. Linedrives should be included in this data. What does Balls In Play data say for ground balls vs fly balls when you have to choose one or the other for each linedrive?

by giantsrainman on Feb 2, 2010 5:18 PM PST up reply actions  

Why should line drives be included in this at all?

Those fall in over 70% of the time, most of the time, so obviously they’re the worst on average.

Aaron King is still my homeboy... iffy mechanics and all

McFAQ for all you newcomers out there.

by baetown415 on Feb 2, 2010 5:31 PM PST up reply actions  

Because The Assumption Is That Pitchers Control HRs, Ks, and BBs Primarily

With a secondary effect on flyballs vs groundballs. You need to elimiate all three items primariy controlled by pitchers to get a good picture of these secondary items. I consider line drives as just two sub sets of fly balls and ground balls. Linedrives that could be fielded by an infielder if he was positioned in it’s path are in essense super ground balls and line drives that can not are in essense super fly balls. Linedrives are an attempt to measure hard contact but they even fail here in that some ground balls (screamers down the line for instense) and some fly balls (HRs for instense) are equally hardhit.

by giantsrainman on Feb 2, 2010 6:16 PM PST up reply actions  

Line drives are very subjective when it comes to recording them…

by Missing Barry on Feb 2, 2010 6:43 PM PST up reply actions  

I kind of just look at groundballs v. non-groundballs

Because then I don’t have to worry about the flyball/fliner/liner classification issues.

And either way, ground balls are favorable to non-groundballs.

Aaron King is still my homeboy... iffy mechanics and all

McFAQ for all you newcomers out there.

by baetown415 on Feb 2, 2010 9:32 PM PST up reply actions  

Yeah, I don’t look at LD’s at all. Ever, really. I mostly just look at GB%, if it’s low, I assume flyball pitcher, if it’s high, I assume GB pitcher. Those are the most accurately recorded batted ball type….

by Missing Barry on Feb 3, 2010 7:25 AM PST up reply actions  

Why in the world would we exclude home runs? They’re the numbers one reason for not wanting fly balls. Asking me to exclude them is just conceding you were wrong. Ground balls are better than fly balls, including in AT&T. Act your age and admit it.

" Because The Assumption Is That Pitchers Control HRs, Ks, and BBs Primarily

With a secondary effect on flyballs vs groundballs. "

This is just false. Nobody claims a secondary effect effect on flyballs and groundballs. If anything the secondary effect is on the HR’s (to an extent). HR’s are, at large, a result of giving up many flyballs (though there are other factors, mainly the home park, and is some cases – more often with relievers than starters – a special ability to avoid turning fly balls into home runs). The top 5 pitchers in GB% last year all had a HR/9 between 0.33 and 0.91. The bottom 5 in GB% all had a HR/9 between 1.11 and 1.58. This is all basic, intuitive stuff. How many sinkerballers do you know who give up many home runs?

I didn’t include line drives for two reasons: first off all, I wasn’t sure if there’s a strong correlation between line drives and fly balls, so I didn’t know if they’re relevant. I also had no doubt that if I would just divide it to “balls on ground” and “balls in air”, you’d say that I’m distorting the numbers by adding hard hit balls to the fly ball group. But if you want to do it that way – be my guest. It would only emphasis how much better ground balls are than balls in the air.

I was promised lasagna.

by Cookyman on Feb 3, 2010 1:55 PM PST up reply actions  

Re Ordering Your Words And You Then You Get It Right
HR’s are, at large, a result of giving up many flyballs
This is just false.

Yes it is true that a ground ball can’t become a HR no matter how hard it is hit but it is also true that weakly hit fly balls can’t become HRs either. There are FB pitchers that are effective at keeping HRs totals down just like there are GB pitchers that are. There are GB pitchers that give up alot of HRs just like there are FB pitchers that do. Being a FB isn’t the main problem, being hit hard is the main problem.

by giantsrainman on Feb 3, 2010 5:47 PM PST up reply actions  

Ah, beautiful out-of-contextness. The full sentence:

HR’s are, at large, a result of giving up many flyballs (though there are other factors, mainly the home park, and is some cases – more often with relievers than starters – a special ability to avoid turning fly balls into home runs)

In conclusion, your comment contradicts nothing. Of course there are exceptions. But there is a strong, clear connection between giving up fly balls and giving up home runs. This connection is the reason why fly balls are much better than ground balls for the hitter. This is backed up both by numbers and by baseball common knowledge (and common sense). You claimed that this isn’t true in AT&T, and I showed it to be false. For the pitcher, ground balls are much much better than fly balls, including in AT&T. This is supported by the fact that every single Giants pitcher from the past 10 years did better on ground balls than on fly balls. You’ve provided absolutely no reason to expect Bumgarner to be any different.

I was promised lasagna.

by Cookyman on Feb 4, 2010 4:26 AM PST up reply actions  

There are FB pitchers that are effective at keeping HRs totals down just like there are GB pitchers that are. There are GB pitchers that give up alot of HRs just like there are FB pitchers that do.

Well that’s why we looked at the end results. And the end results still say GB’s > FB’s…

by Missing Barry on Feb 4, 2010 9:07 AM PST up reply actions  

A View From Tom Tango
As we know, the run value of a groundball (in MLB) is about the same as the run value of a flyball (if we exclude HR).

Link

by giantsrainman on Feb 2, 2010 8:00 PM PST up reply actions  

Yeah, that Tango quote is used in a completely different context. For our purposes, there’s no reason to exclude HR’s, and once you include them, it’s very much in GB’s favor.

by Missing Barry on Feb 3, 2010 7:27 AM PST up reply actions  

Overall, it is still much much better to get ground balls.

Cookyman’s point. When the ball is put into play, it is better to be on the ground than in the air. In the air, HR’s happen, and should be included in the run value. On the ground, they don’t happen, and once you factor in HR’s, on the ground is clearly better than in the air.

by Missing Barry on Feb 3, 2010 10:14 AM PST up reply actions  

HRs Don't Just Happen

I am not and never have been a believer that there is one magical percentage of HRs/FBs that one should expect for all pitchers. I believe some FB pitchers can hold HRs down just as well as many GB pitchers can. This is why I believe Tom Tango removes them from this discussion as well.

by giantsrainman on Feb 3, 2010 5:38 PM PST up reply actions  

LOL Bhaakon

You expect GRM to share evidence?

To borrow from Lars:

Aaron King is still my homeboy... iffy mechanics and all

McFAQ for all you newcomers out there.

by baetown415 on Feb 3, 2010 7:41 PM PST up reply actions  

I Just Have A Different Perspective

I don’t believe degree of trajectory should define the brackets. I believe degree of contact should. In my book home runs should be combined with line drives and screaming ground balls into the hard contact bracket. The next step down should be average contact (average ground balls finding or not finding holes and average fly balls finding or not finding gaps). These leaves weak contact as the final bracket (pop ups and squibs). The goal of a pitcher is to avoid the first and maximize the second. This can be done regardless of the trajectory the pitchers stuff and style is most suited to inducing.

by giantsrainman on Feb 3, 2010 8:01 PM PST up reply actions  

Is it wrong that this actually makes sense to me?

I'm thinking but nothing's happening.

by JRPhillips on Feb 4, 2010 10:31 AM PST up reply actions  

And that’s fine.

Every single Giants pitcher does better on ground balls than on fly balls

Matt Cain:
.216/.216/.240 on ground balls.
.192/.188/.495 on fly balls.

Tim Lincecum:
230/.230 /.258 on ground balls.
192/.188/.495 on fly balls.

Jonathan Sanchez:
.261/.261/.289 on ground balls.
.218/.214/.582 on fly balls.

This isn’t assuming some standardized HR/FB. This is just taking the end results of balls hit on the ground vs. balls hit in the air. If a guy is good or bad at letting up HR’s, it will be included in this. If our park helps prevent HR’s making certain pitchers a better fit, it will be included in this. And this still suggests that more GB’s are better.

by Missing Barry on Feb 4, 2010 9:09 AM PST up reply actions  

I Don't Consider HRs Flyballs - They Are Hard Contact

Just as linedrives and screaming grounders are. Trajectory should not be determining the brackets we look at.

by giantsrainman on Feb 4, 2010 11:55 AM PST up reply actions  

So, in your mind, GB% is completely irrelevant to a pitcher’s performance, aside from maybe getting a few more double plays? You realize that contradicts both common knowledge and a hell lot of evidence? And is backed up by nothing?

I was promised lasagna.

by Cookyman on Feb 4, 2010 1:18 PM PST up reply actions  

Yes And No

It is so called common knowledge because of the original assumption to bracket things by trajectory. Nobody has yet measured things as I am suggesting would be a more reasonable why to bracket things.

by giantsrainman on Feb 4, 2010 1:43 PM PST up reply actions  

I Am Questioning This Original Assumption

Not the data measured under this assumption. I think it just might be wrong to look first a trajectory as opposed to look first at degree of contact.

by giantsrainman on Feb 4, 2010 1:45 PM PST up reply actions  

But if there’s a clear correlation between FB% and home runs, I don’t see how it matters. The list of extreme fly ball pitchers who have a low HR/ball in play rate is very short. As is the list of ground ball pitchers who have a high HR/ball in play rate. In general, pitchers who give up more fly balls, give up more home runs. You’ve given absolutely no reason for us to doubt this.

Every single pitcher in modern baseball history did better on ground balls than on fly balls. You’ve given no reason to believe Bumgarner will be different. Your claim that, in AT&T, fly balls are better, has been completely refuted (and you showed no evidence to back it up in the first place). It’s not rare for you to refuse to admit a mistake, but I’ve never seen you take it to this level.

I was promised lasagna.

by Cookyman on Feb 4, 2010 2:15 PM PST up reply actions  

I Think There Is Just As Much Correlation Between Hard Contact And HRs

More importantly I think hard contact has more causation relative to HRs then Flyballs do.

by giantsrainman on Feb 4, 2010 6:49 PM PST up reply actions  

Cool

Doesn’t contradict anything. Fly balls pitchers allow more home runs.

I was promised lasagna.

by Cookyman on Feb 5, 2010 2:25 AM PST up reply actions  

Who are your “other starters” – are you expecting us to sign a Free Agent? If not, that seems pretty optimistic, especially since that includes innings going to emergency AAA guys.

Lincecum seems about right, though I wouldn’t be surprised if he beats that.

I’d say both Cain’s and Sanchez look too optimistic to me (especially 195 IP from Sanchez). Both are possible, but I wouldn’t count on them

And I really wouldn’t count on our top 4 relievers all reaching 75 IP. Does that ever happen? None of them did it last year (Wilson and Medders came close). Also, not a big issue, but I don’t think Romo will keep on giving up so few home runs. He’s an extreeeme flyball pitcher.

I don’t know what to expect from the rest of the bullpen. I guess 5.25 ERA sounds as good as anything.

I think that adds up to about 30 runs. So that would be the difference between our projections.

I was promised lasagna.

by Cookyman on Feb 1, 2010 9:13 AM PST reply actions  

I don’t think Sanchez’s rate stats are crazy optimistic, they’re within spitting distance of what he did last year. But I do agree that I would expect closer 170 IP than 200.

FU, FO

by oldjacket on Feb 1, 2010 9:22 AM PST up reply actions  

It’s the ERA in Cain’s case and the IP in Sanchez’s.

I was promised lasagna.

by Cookyman on Feb 1, 2010 9:27 AM PST up reply actions  

In My View Sanchez Is Ready To Be Consistent Enough To Get To 200 Innings (Health Has Never Been An Issue)

The projection tool I used from Fangraphs in essence forced me to choose between 45 innings and 75 innings for the four individual relievers I rated so I went with 75. I do realize that this is a bit high which is why I intentionally made the 153 innings by the others look worse then they are actually likely to be to compensate for too many good innings from the first four.

Finally, to increase the runs the Giants allow by another 30 runs on top of the 37 I have already increased it from last year is being pessimistic for pessimistic’s sake in my view and as such is just not justified.

by giantsrainman on Feb 1, 2010 3:04 PM PST up reply actions  

I'm going to ignore the last sentence

That’s weird about the bullpen’s IP. Why can’t you project 60?

Last year, there wasn’t a single NL team that had 3 pitchers reach 195 IP. You’re projecting us to have 4. I doubt that’s even happened in the past 5 years.

I was promised lasagna.

by Cookyman on Feb 1, 2010 3:57 PM PST up reply actions  

Rare Is Not Always The Same As Unlikely.

In my view the Giants start with two as virtually givens (Lincecum and Cain) and a third damn close as Zito has not pitched less then 180 innings since his rookie year. Again, the choices were ranges and I just could not justify putting any of these three in the 150 to 179 range which would be averaged to 165. Sanchez was the only one I considered for the 150-179 range and this would fit his two years as a full time MLB starting pitcher. But, I felt uncomfortable doing so as I am convinced he made a breakthrough last year and is now ready to get to 200.

by giantsrainman on Feb 1, 2010 4:25 PM PST up reply actions  

I didn’t say you made a wrong choice. Just that the result is too optimistic.

I was promised lasagna.

by Cookyman on Feb 1, 2010 4:34 PM PST up reply actions  

That innings thing does seem to make it difficult. It really doesn’t seem like it would be that hard to let you pick the actual number, I wonder why they do that…

by Missing Barry on Feb 1, 2010 8:31 PM PST up reply actions  

Because They Weight And Arerage All The Different Ranges Selected By Multiple Fans

For my purposes I don’t see that much need to try and be any more accurate on individual players and thus prefer to make my team adjustments as needed in my “Other Starters” and Other Relievers" so that I don’t have to edit them all.

by giantsrainman on Feb 1, 2010 9:22 PM PST up reply actions  

ha just read this!

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 1, 2010 4:30 PM PST up reply actions  

i'm curious GRM

 in wilson and romo, both with 75 ip and romo with a better whip, how is is era 1/2 a run higher with giving up the same hr ?

by giantdonkey on Feb 1, 2010 9:47 AM PST reply actions  

But, uh, his WHIP is still higher than Romo’s…

Barry Zito: Mike Hampton with a guitar

by JakeS on Feb 1, 2010 9:41 PM PST up reply actions  

And, If You Look There Are Other Examples

I show Sanchez and Zito with the same WHIP but Sanchez with a better ERA due to fewer hits and bombs. I show Romo and Affeldt with the same ERA, hits and HRs even though Affeldt has a higher WHIP due to more walks because of Affeldt’s ability to induce DPs.

by giantsrainman on Feb 1, 2010 9:55 PM PST up reply actions  

These projections must be based off of Bochy calling pitches from the bench. No half-wit rookie catcher could ever lead a pitching staff like this!

/auto-defenestrates

Dear Internet,
Please fire Brian Sabean.
Signed,
Me

FREE KEVIN FRANDSEN!!! Member of the Frandsen 5% Club.

by Uribe nee Gonzalez on Feb 1, 2010 12:04 PM PST reply actions  

90 comments, 105 of them from rainman....

just sayin’

Brian Sabean figures that if he buys enough bottles, one of them is bound to have lightning in it.

by jasomack on Feb 1, 2010 7:51 PM PST reply actions  

That’s the kind of math that belongs in a thread like this.

Barry Zito: Mike Hampton with a guitar

by JakeS on Feb 1, 2010 9:41 PM PST up reply actions  

how did you decide

that the top 4 relievers would all give up 5 HR in 75 IP. 7.5 HR would be ~league average, or xFIP approximation.

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 2, 2010 8:41 AM PST reply actions  

That's just the FanGraphs tool

It doesn’t give you more specific options.

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 10:06 AM PST up reply actions  

huh?

so GRM did NOT project them to pitch 75 innings with 5 HR? Fangraphs did based on some input? Like ERA K/9 BB/9?

TEMPORARY SIG AWAITING FINAL SABEAN APOCALYPSE
-------
PARPG- Indy post-apocalyptic roleplaying game in development

by zenbitz on Feb 2, 2010 10:37 AM PST up reply actions  

Yeah, I brought this up in the beginning — he’s just aping Fangraphs fan tool thingy with his own inputs.

#1 FanShot Champion

by xanthan on Feb 2, 2010 10:45 AM PST up reply actions  

So, for example

If I want to project Brian Wilson with their tool, I get to choose one of the following for each category:

IP: 1-29, 30-59, 60-89, etc.
Role: Starter, MR, SU, CL, etc.
Wins: 0-2, 3-5, 6-8, etc.
Losses: Same as wins
ERA: 2.00-2.49, 2.50-2.99, 3.00-3.49, etc.
K/9: 3.0-3.9, 4.0-4.9, etc.
BB/9: 1.0-1.4, 1.5-1.9, etc.
HR/9: 0-.24, .25-.49, etc.

So, if you choose Wins: 3-5, you are actually projecting 4 wins, which is why all the relievers up there have 4 wins.

I don’t think it’s a good tool for one person to make projections like this. I think it’s use comes from many people making the projection on one player and averaging out the results.

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 10:49 AM PST up reply actions  

Yes!, Not designed for single projections.

I don’t think it’s a good tool for one person to make projections like this. I think it’s use comes from many people making the projection on one player and averaging out the results.

Exactly right. It was not designed for single projections. Next year though, we may have to change things.

by dkappelman on Feb 2, 2010 4:13 PM PST reply actions  

While You Are At It

Can you change how you use ERA to determine hits? I had to edit hits for ERAs of 3.75 and above way down after copy and pasting them over here to make them realistic. Your system calculated Barry Zito giving up 230 hits in 195 innings because I selected an ERA range of 4.00 to 4.49, Other Starters (edited from Randy Johnson after copy and paste) giving up 218 hits in 165 innings because I selected an ERA range of 4.50 to 4.99, and Other Relievers (edited from Merkin Valdez after copy and paste) giving up 198 hits in 135 innings (I also edited this to 153 innings after copy and paste) because I selected an ERA range of 5.00 to 5.49. All of these calculated hits are way too high!

by giantsrainman on Feb 2, 2010 5:01 PM PST up reply actions  

Thanks for running such an amazing site!

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 5:41 PM PST up reply actions  

lol.

David Appleman: “Ur doin it wrong”

Also, <3 Fangraphs!

#1 FanShot Champion

by xanthan on Feb 2, 2010 5:44 PM PST up reply actions  

*Appelman, too. But I bet you get that a lot.

#1 FanShot Champion

by xanthan on Feb 2, 2010 5:45 PM PST up reply actions  

Second comment ever on McC

Wonder what the first is.

BTW, any news re: catching defense that I’ve heard you all were trying to implement this offseason?

Aaron King is still my homeboy... iffy mechanics and all

McFAQ for all you newcomers out there.

by baetown415 on Feb 2, 2010 5:53 PM PST up reply actions  

POWER SEARCH

I HAVE THE POWER

Juan Carlos Perez, please start hitting.

by marcello on Feb 2, 2010 5:56 PM PST up reply actions  

This Thread Is Insane!

I feel tired reading it.

Giant Dirtbags: John Bowker, Steve Hammond. MIA List: Todd Jennings, Brian Anderson
Jeremy Affeldt induces DP's

by Giant among Angels on Feb 3, 2010 8:17 PM PST reply actions  

Let’s discuss Fred Lewis’ defense now..

#1 FanShot Champion

by xanthan on Feb 4, 2010 5:31 AM PST up reply actions  

I think we need someone to sum up what we learned in this post, South Park style.

FU, FO

by oldjacket on Feb 4, 2010 3:43 PM PST reply actions  

Re-reading this thread, I’m still baffled by what GRM tried to achieve here. We’re a playoff team according to his projections – but the projections are based on the assumption that we’re a playoff team ( “This is the only way they Giants are going to make the playoffs when they do so”, " For Non $120M+ Payroll Teams That Make The Playoffs This Is Rarely True").

So, 250 comments later, I still don’t know what’s going on here.

I was promised lasagna.

by Cookyman on Feb 5, 2010 2:55 AM PST up reply actions  

So, 250 comments later, I still don’t know what’s going on here.

…the aristocrats!

#1 FanShot Champion

by xanthan on Feb 5, 2010 3:37 AM PST up reply actions  

Comments For This Post Are Closed


User Tools

Welcome to the SB Nation blog about San Francisco Giants.
Yahoo_full_count

Manager

174246766_ea2fd78204_small Grant Brisbee

Moderators

Sbzito_small Natto

Fawlty_small WalrusMan

Goofus_small Goofus

Howtheyscoredcat_small howtheyscored

Det_7193_small jponry

Authors

09_small JT Jordan

Small steve S

E6dmccicon_small Every6thDay