I demand more pie charts.

So Bret “I have an OPS under .400” Gardner is a better offensive force than Melky this year?  (pro-rated to 435 plate appearances Gardner would check in at -7.8)

I want to cry.

I only had to press the Page Down button seven times to see the comments section.  I’d be willing to go nine or ten times if pie charts were included.

I’m holding out for downloadable 3D representations of the Yankees season.

FireRobinsonCano.com

Apropos of nothing, we just had a friggin’ monsoon blow through midtown Manahattan.  Perhaps it has washed away the crappy Yankee play?

I would be interested to see how the Yankees of the non-championship post O’Neil years (2002-2008) compare to the Yankees we miss and love (1996-2001) in terms of clutchness.

Apropos of nothing, we just had a friggin’ monsoon blow through midtown Manahattan.  Perhaps it has washed away the crappy Yankee play?

Just hit us here in jersey a few min ago.

Not that I am whole heartedly endorsing this “clutch” stat, even though I do, as SG put it “think pressure can have some impact…”

I do agree that it is in fact a stat, with values that can be measured.  When I see a chart like Clutch expressed in terms of runs above and below average, and see that we’re -26 below average… I can associate that with the play that we’ve all been watching this season.  Now, weather this should be called “clutch” or something else loosely defined as ability to push in the RISP…I suppose that’s up to one’s own judgement.

PS…It’s now sunny and beauitful outside.

7: Are you sure you want to go down that road? Surely there must already be enough to make you depressed about this team. 

It would be interesting to see, though.

I would be interested to see how the Yankees of the non-championship post O’Neil years (2002-2008) compare to the Yankees we miss and love (1996-2001) in terms of clutchness.

HBT has data back through 2004.  For the 2002-2003 I’d have to compute it myself.  Give me a bit and I’ll see how hard that is to do.

For 2004-2007, here are the team totals:

The Yankees have been so unclutch, they’ve all forgotten how to drive standard.

Ok, thanks to David Pinto’s valuable Day by Day database this is pretty to easy to calculate back through 2000.

The Yankees are so unclutch they strand runners even when the bases are empty.

I am very pleased to note that the 2000-2001 teams were unclutch.  Can I please not have to read about the O’Neill-Tino-Brosius Axis of True Yankeedom anymore?

there is no question that this team has sucked with RISP.  but i don’t really know if this definition is really that great, since it is certainly possible to do “clutch” things without men in scoring position.

for example, on Saturday, both A-Rod and Giambi’s HRs were definitely “clutch”.  6th inning of a scoreless game?  those were HUGE HRs (of course, knowing now how the game turned out, it is forgotten how big those HRs were).

A-Rod’s double yesterday in the late innings was also “clutch” (again, knowing he was “out” stealing 3B kindof leaves a bad taste). 

what’s my point?  i don’t know.

Yeah, clutch is probably not the best name for this stat.  Something like RISP and ROB +/- or something would be better.

but i don’t really know if this definition is really that great, since it is certainly possible to do “clutch” things without men in scoring position.

I definitely agree, which is why it is hard to put a number on how “clutch” a player is.

Also, I think there have definitely been circumstances where the player at the plate has been about as clutch as he can (minus hitting a HR), and they haven’t gotten anything out of it.  Line drive right at a defender, great play in the outfield, I think there have even been some double-plays where off the bat you’re thinking, “that’s at least one, maybe two runs”.  Just one of those years that the breaks haven’t gone our way.  Now we can just hope there’s some retributation in the last 6 weeks, to get into the playoffs.

SG- What’s your take on WPA?

off topic:  Posada’s injury got me thinking.  If in fact his injury is bad enough that he can’t catch anymore, is there any chance he would retire and/or the Yankees would get insurance compensation for him?  Furthermore, why is there the assumption that Posada would then slot in at DH or 1st.  Can we really expect his bat to play at those positions?  I know his contract is a lot of money, but it’s already a sunk cost.  It would be crazy to turn away better players just b/c we already have Posada to play 1st base poorly.

Just one of those years that the breaks haven’t gone our way.  Now we can just hope there’s some retributation in the last 6 weeks, to get into the playoffs.

no doubt. 

perfect example:  i thought A-Rod had a huge jump on that steal yesterday.  but the pitch just HAPPENED to be up and in, a perfect spot to throw to 3B from. 

look at the location of every pitch in the AB and the location of the pitch he happened to steal on:

http://mlb.mlb.com/mlb/gameday/index.jsp?gid=2008_08_10_nyamlb_anamlb_1&mode=gameday

there was no REASON for the pitch to be in that spot.  but it just was.  if the pitch is anywhere else, A-Rod steals the base easily.

same thing with the ball Jeter hit the previous inning.  with a little luck, it’s over Kendrick’s head.  same with the ball he hit in the game against Texas where Hamilton made the sliding catch.

i’m not saying luck is the reason this team is where they are.  it isn’t.  they’ve mostly made their own bad luck.

but man, it does seem like in the last week, nothing has gone right.  it’s as if they used up all their luck for the month in last sunday’s game.

Reading that definition of clutch and A-Rod and Giambi’s numbers reminded me of the 3rd inning of Friday night’s game (http://www.fangraphs.com/plays.aspx?date=2008-08-08&team=Yankees&dh=0).  There were runners on 2nd and 3rd with 2 out when A-Rod came up.  He was intentionally walked.  Giambi then worked a walk for an RBI.  Neither of them got any clutch credit, yet those walks were worth something—if less than what a hit would have been.

Say that a walk with the bases loaded is 60% as good as a single, then a player with 20 PA in that situation who gets 2 singles and 5 walks (BA 0.133) has done as much for the team as one who got 5 singles and 0 walks (BA 0.250), yet clutch clobbers the first player.

How easy would it be to find A-Rod and Giambi’s walk rates with RISP?  It wouldn’t surprise me at all to learn that they are being pitched around to some degree, and if their walk rates turn out to be very high, that could explain some of the unclutchiness.  For good measure, if there is a psychological effect on the players from fan perceptions of clutchiness, I’d like at least informed Yankee fans to be aware of walks with RISP so that we don’t unintentionally encourage players to press precisely when they are at greatest risk of being pitched around.

Can we really expect his bat to play at those positions?

his bat plays just fine at 1B if he simply hits to his career average.

league average at 1B is .263/.345/.430 this year

his typical season is like a positional adjusted OPS+ of about 121.

How easy would it be to find A-Rod and Giambi’s walk rates with RISP?

A-Rod is hitting .239 with RISP but his OBP is .399

Giambi is hitting .204 with a .358 OBP.

there may be something to the idea that they aren’t getting much to hit.

it may be less true now with Nady, Pudge, good Robbie, etc.  but earlier in the season, i don’t doubt it.  once you got past Giambi in the lineup, there was absolutely no one to hurt you.  this went on for a month or two.  Molina, Melky, Gardner?  bad Robbie?

What’s your take on WPA?

It measures something I guess, but it’s no better than RBI or W-L to me, since it’s so team-dependent and timing dependent.  I can’t get behind a stat that says a HR is worth more depending on when it’s hit.

I like the graphs that show win expectancy over the course of a game on a team level, but I don’t care for WPA as a basis for ranking individual players.

A lot of people whose opinions I respect seem to like it, especially when dividing WPA by LI (leverage index) which should help to smooth out the team specific differences, but I honestly never even look at it.

How easy would it be to find A-Rod and Giambi’s walk rates with RISP?

Very.

If there is such a thing as unclutch, that 2004 Yankee team was certainly it-  Especially in games 4-7 of the ALCS.  I agree that “clutchness” is overblown but I also think that some players seem to change their approach in an obvious manner in certain situations.  ARod is most guilty of this in my mind.  I hope Madonna can brainwash him w/Kabbalah to just be ARod no matter what situation he comes up in.

Thank you, SG.

What about looking at WPA for all situations of above a certain leverage index, say 1.5 or so?  It seems like this would be a fairly decent way to measure “clutch”, but you would also need to think about the overall ability of the player if you wanted to talk about someone “raising their game” at tense moments.  To play devil’s advocate, I have no doubt that ARod hits better than Eckstein in every situation, but maybe little Davy grits out more productivity compared to his baseline when it matters.

I think “clutch” is basically just small sample size issues for the most part, especially over the course of a single season.  It would be interesting to look at a whole career though, because I’d bet you’d have enough AB’s in high leverage situations that at least some of the data would be statistically significant.

Thinking about it a little more, just taking straight WPA would skew the results to just a few AB’s with really high LI, it might be better to look at the WPA/LI, but restrict it to relatively high leverage situations.

Thurman Munson hit .302 for his career w/RISP.

With men on base (MOB), Munson batted .307.

Carlton Fisk: RISP .270, MOB .274

Gary Carter: RISP .271, MOB .272

Johnny Bench: RISP .262, MOB .273

Yogi Berra: RISP .265, MOB .264

Clutchiness is next to truthiness

“A-Freud”: RISP .303, MOB .311

“Clutchness and the 2008 Yankees”

Can those two concatenated phrases coexist in the same thread title?

so dunn got traded…i cannot recall a bigger player getting traded after the NWTD…can anybody help me?

Nice one, Yup.

</i>i cannot recall a bigger player getting traded after the NWTD…can anybody help me? </i.

oh, you have obviously forgotten the infamous Gabe White/Felix Heredia acquisition of 2003.

Canseco, 2000.

Randy Myers, 1998.  (He was toast but he was still a big-name closer, after all.)

Also, The Yankees and Mariners swapped Jeff Nelson and Armando Benitez after the deadline in, what, 2003?

Of course these are not necessarily bigger names than Dunn, but certainly in the same echelon.

Dunn is the biggest name I can think of.

The dude has a 900 OPS!

Jayson Stark: Of course! My “runs produced” definition is RBIs plus runs scored, then subtract homers.

Now, compare the level of analysis given by Stark to that found here, and then remind me which one gets paid to do it.

Jayson Stark: Of course! My “runs produced” definition is RBIs plus runs scored, then subtract homers.

Jayson Stark is not a good baseball analyst.

A-Rod is hitting .239 with RISP but his OBP is .399

Giambi is hitting .204 with a .358 OBP.

For Giambi, the walk rate isn’t much changed (overall .252 BA .399 OBP in 2008, for .147 ISOD overall compared to .154 with RISP), so he might not be getting pitched around all that much.

A-Rod, on the other hand has a 2008 ISOD of .087 overall, but .160 with RISP) small sample size and all, that’s statistically significant (he has 109 AB with RISP, so if the true value of his ISOD with RISP is equal to his season value of .087 or his career value of .082, that would make the .160 well over 2 sigma).

To dot the i’s and cross the t’s, if we assume that we know the true value for ISOD and it’s .082, then the standard error for a sample is the single trial SD (sqrt(0.082*(1-0.082)) = 0.274) divided by sqrt(109 - 1), yielding a SE of .026 or .027 depending on whether we use this year’s ISOD or his career number.

That gives us a z score of 2.7 against this year’s ISOD (p=0.35%) or 2.9 against his carreer ISOD (p=0.19%).  I wouldn’t trust these estimates exactly (since I didn’t account for uncertainly in our estimate of his “true” ISOD and such), but I think it’s fair to say that there’s less than a 1% chance that his walk rate with RISP is the same as his overall walk rate based on what we’ve seen so far.  Since the extra walks are coming at the expense of hits rather than in addition to them (probably because he’s being pitched around), I see this as a problem.

I guess that the Yankees are already working on this problem, since the lineup after A-Rod is a lot stronger than it used to be.  Assuming that it’s still a problem, the single biggest tactical change that I would make to try to get A-Rod better pitches to hit, would be to tell Jeter to stop bunting.  Imagine that there are runners on first and 2nd with no outs when Jeter comes up (the prototype bunt situation).  If Jeter sacrifices successfully and there are now runners on 2nd and 3rd with 1 out, Abreu is likely to get a good opportunity to hit.  If he fails, though, A-Rod is often going to be pitched around or walked.  If Jeter could stop GIDPing, his career .386 OBP (and to a lesser degree his season OBP of .346) would argue strongly for trying to get on base rather than bunting in those situations.

I can’t get behind a stat that says a HR is worth more depending on when it’s hit.

While this analysis (which, as always, is great SG, thanks) doesn’t do that the way WPA does, if we take it and weight it more in our minds so we can justify why the Yankees aren’t playing as well as we think they should, isn’t pretty similar?

Think about it this way: the Yankees are 4 games behind the Red Sox for the WC lead. I think we would all feel a lot better if they were tied for the WC.

So, can you think of 4 games where 1 hit could have resulted in a win (for example, the 1-0 loss to Anaheim in NY)? I’m sure it wouldn’t be too hard to find that. The Yankees have combined for over 4000 AB’s this year - so the 4 AB’s that you change that result in wins represent less than .1% of their total AB’s.

Like yup, I don’t know what my point is.  I still have a hard time believing that some players get worse and some get better in certain situations. But that’s not really the argument here.. what we’re looking at is what’s brought us to this point and if there is any hope for the future. /ramble

I think the thing to do is to think about what you want to be measuring when thinking about WPA or other offensive stats.  If you’re retrospectively looking at what happened in a season or a game, for the MVP say, I think WPA is a good stat to at least consider. So, for looking at “clutchness”, it seems like WPA is a good place to start.  Past performance in these stats is no indication of future returns, to parrot an investment commercial.

If you’re looking at who you would want on your team for next year, I think a context-neutral stat is going to be better.  All other things being equal, I’ll take the player with the higher OPS going into next season.

I still have a hard time believing that some players get worse and some get better in certain situations.

I agree; however, you can never truly know.  No player is ever going to admit to the press that he chokes under pressure.  And no player has the stones to say he’s better in certain situations than others - especially when comparing “clutch” versus “non-clutch” situations.  If that were the case, the media would jump on the player for dogging it during non-clutch situations.

While it’s quite hard, if not impossible, to prove using stats and even harder for a player to admit, I do think some players respond to pressure better than others.  I believe this because it is a part of life: some people respond to pressure better than others, whether it be live-saving or life-threatening situations, speaking in public, or playing sports.

Misery loves company Dept:Mets blow 5-1 lead, 5-4 after 8, losing to Bucs 7-5

Is anyone else a little disapointed that Crawford got injured?  At this point I’ve given up on catching the Rays, even if they fall out of the division lead, so I’d rather see them do well than not.

Jayson Stark: Of course! My “runs produced” definition is RBIs plus runs scored, then subtract homers.

I like that he subtracts out home runs—wouldn’t want to count those runs twice! That’s some top shelf logic right there!!

That’s some top shelf logic right there!!

Yep, strikeout = leadoff triple

To be fair to Stark, this formula has been used by many people for many years.  I.e. he inherited rather than invented the idiocy.

To be fair to Stark, this formula has been used by many people for many years.  I.e. he inherited rather than invented the idiocy.

Are you saying that makes it OK?

Yeah right.  No, I’m saying something along the lines of don’t shoot the messenger, or history is a nightmare from which I’m trying to awake, or something totally different from those things.

Abraham reports Joba to start playing catch Wednesday.  Perhaps not all is lost.

Jayson Stark makes me nauseous. He’s a egomaniacal douchebag, not to mention an absolute hack writer. His pieces range anywhere from pseudo-scientific garbage that he’s ripped off and bastardized (see above) to completely nonsensical. Not to mention, he writes on about an 8th-grade level.

Sorry, I just can’t stand Jayson Stark. More generally, ESPN’s quality of articles has absolutely plummeted over the past few years.

ESPN’s “quality” was never high.

Jayson Stark makes me nauseous. He’s a egomaniacal douchebag, not to mention an absolute hack writer. His pieces range anywhere from pseudo-scientific garbage that he’s ripped off and bastardized (see above) to completely nonsensical. Not to mention, he writes on about an 8th-grade level.

Sorry, I just can’t stand Jayson Stark. More generally, ESPN’s quality of articles has absolutely plummeted over the past few years.

I couldn’t have better described my own feelings. I have never seen someone who knows so little so full of himself.

Is anyone else a little disapointed that Crawford got injured?

If it means Rocco Baldelli gets some playing time, I think it’ll be ok.  I’d like to the Yankees to jump all over Rocco in the offseason, if for no other reason than he is high enough ceiling that throwing a few million at him to see if he can come back strong is worth it. Not sure what he would demand, but I think he this is his age 26 season.

I still have a hard time believing that some players get worse and some get better in certain situations. But that’s not really the argument here.. what we’re looking at is what’s brought us to this point and if there is any hope for the future. /ramble

Right.  This isn’t an analysis of a clutch skill or ability, it’s merely a look at the value of the AL teams at hitting with RISP and ROB to this point in the season.  If anything, we should expect all teams to regress towards average going forward, absent any knowledge of a particular clutch skill.  So if you want to be hopeful, maybe that’s something to look forward to for the Yankees.

What’s scary is Tampa shows as the second worst team with runners on by this measure.

What do you think are the chances of the Yankees signing Manny after this season? 5%? 10%?

0%.

No Damon in the lineup? Sort of ironic that it comes when this post comes out.

The team isn’t ‘clutch’; they can’t hit with RISP.

Damon is the most ‘clutch’ on the team.

They’re just coming off the a kick sand in your face sweep by the Angels.

So what does Girardi do? Give Damon the day off!

Priceless.

Quick question, this appears to reward positive “clutch” at 2 to 1 ratio and penalizes negative at the same, wouldn’t these clutch runs already be part of the pBRAR totals?

What is the healing process on a right wrist fracture?

That sounds really bad, but the Rays seem to be acting like Longoria won’t be out too long.

How does that work?

If the Rays report injuries anything like the Yankees do, an initial indication like “fractured wrist” means that Longoria probably died in a fiery car explosion.

Quick question, this appears to reward positive “clutch” at 2 to 1 ratio and penalizes negative at the same, wouldn’t these clutch runs already be part of the pBRAR totals?

I don’t think that the clutch runs are part of the pBRAR.  The way I interpret this is that we figure out the batting runs value of a player’s offense in a neutral context, then adjust that up or down based on this clutch number.  I assume that the 2 to 1 ratio is how James corrected for giving positive or negative credit to batting events that helped or hurt your team’s chances at actually scoring in the context in which they happened.

The original “Clutch” addition to runs created was done due to a problem with James’s original runs created formula where team totals didn’t line up with their actual runs scored.  By factoring in hitting with RISP and HRs with ROB James was able to improve the accuracy of his runs created formula on a team-level, but it’s certainly debatable if this works on an individual player level.

I’m kind of with IllusionofTime on this.  I don’t know if this stat is particularly great, but it does seem to line up with our perception that the Yankees have not hit well in key situations this year.

The Yankees are as far back in the loss column as the Giants.

I’m now 10% more depressed.

SG Thanks for the clarification