Archive for June, 2009

WPA Inquirer

Dave Studeman introduces a new WPA tool at The Hardball Times. Very cool.

It gives you the win probability for any situation in a game, and you can also adjust for run environment. The important thing to remember is that the numbers that get spit out are overall averages; you have to keep in mind the specific situation (pitcher, hitter, on deck hitter, base runners, etc.). Also, with any sac bunt analysis, you have to remember that many sac bunt attempts turn into hits, errors, double plays, etc, which changes the numbers.

Anyway, fun stuff. I’m sure we’ll be using it in the future.

Hey, this is something different

A mainstream media vs. mainstream media debate at the U-T. Radio vs. print.

(h/t: Hank’s Padre Discussion)

Gonzalez is good

Haven’t watched enough Padres games this year, but I’m tuning in tonight. Some random observations:

Adrian Gonzalez is pretty good. He turned on a high fastball and hit it way out to left. It’s remarkable how he got his hands through on that pitch and got the good part of the bat on it. Went to the cage today where I struggled to do that on 70 mph fastballs that I knew were coming …

After that, Adrian made a nice diving catch. He’s having a nice year.

Blanco’s driving the ball nicely tonight. Hit a pretty good shot out to right center that was tracked down by Ichiro. Then he lined a double down the left field line.

Everth Cabrera knocked ‘em in, and stole second and third easily. I’m not a huge fan of stealing third with two outs, but if you know you can make it, then I’m for it. It wasn’t very close, so I’ll assume Cabrera knew he could make it (at least a high percentage of the time).

Eckstein squirted a single past Russ Branyan at first, and the Pads lead 4-1. These guys are 30-38? Come on.

Gwynn

After struggling in stints in three seasons with Milwaukee, hitting .248/.300/.298, Tony Gwynn Jr. is dealt to the Padres where he hits .348 in his first 27 games. What are the chances that the son of the best player in Padres’ history manages to not only have enough talent to make the big leagues, but also somehow finds his way onto his father’s old team? And not only that, but he hits like his dad, at least results-wise, in his Padres debut.

You can’t make this stuff up. It’s the kind of story that any sportswriter loves, and let’s face it, as sports fans, it’s the stuff we love too. The story line is set up, but in the Padres front office, there are more important issues regarding Gwynn (though I’m sure they are too enjoying his early success). Namely, how will he perform in the future and what is his future role on this team?

For this season, it appears as if Gwynn has earned himself a starting spot in centerfield. He truly has played excellent baseball, outside of the base running issues (which aren’t a huge deal in the end, anyway). His .348 average hasn’t been empty, as he’s walked in 14% of his plate appearances. The power really isn’t there, but he’s already almost matched his extra base total with the Brewers (he’s one shy in 157 less PAs). His fielding in center has been average, according to UZR. He’s already racked up 1.1 Wins Above Replacement, whereas in his three years with the Brew Crew his performance was right at replacement level.

We have a couple of reasons to doubt that this performance will persist, however; that this, or something close to it, is what we can expect from Gwynn in the future.

Sample Size

It just isn’t many games. Observed performance can really fluctuate over such a small amount of games. Gwynn has only played like 1/5 of an entire season (this year). Even if he played a full season at this level, we’d expect quite a bit less going forward – and he’s not even close to that point yet.

Regression

Again from saber-school 101 — If we didn’t know anything about this guy as a player, we’d expect him to come back down to earth as he accumulates playing time. This is basically the idea that all extreme performance, good or bad, will tend to converge back toward average in time. These first two are clearly related. The smaller the sample, the more we “regress” toward the mean.

‘Peripherals’

There are some numbers we can glance at that may explain some of Gwynn’s hot start. For instance, Tony’s Batting Average on Balls in Play (BABiP) sits at .438. Hitters do indeed have some control over their BABiP’s (much more so than pitchers), but not to this degree — no qualifier was above .400 last year, for instance. Gwynn’s line drive percentage has been a remarkably high 31% this year. That combined with solid speed will likely make for a nice BABiP. The issue, however, is the sustainability of that line drive rate ….

Past performance

Unlike the ‘dummy player’ suggested above, we do know something about Gwynn, as he’s racked up 371 major league PAs and 2686 in the minors. So far, he’s hit .271/.341/.344 in the majors and .275/.348/.345 in the minors. Let’s take a look at how some familiar projection systems viewed Jr. before this season:

System AVG OBP SLG
CHONE .266 .334 .342
Marcel .258 .325 .362
Oliver .252 .308 .320
ZiPS .257 .313 .314

 

Yeah, not great, and they are of course based on his relatively pedestrian track record, both in in the minors and majors (well, except Marcel, who doesn’t consider the minors). An updated ZiPS version, that incorporates this year’s stats, projects Gwynn at .265/.324/.330 the rest of the way. The effect of this season is there, but again, because it’s in such a small amount of games, it does not dramatically change Gwynn’s projected ‘true talent.’

*If* this is the real Gwynn, well, he’s still a decent player. He isn’t a guy you really want to build around, however, or even a guy you want to entrench into a position for a few years. He’s somebody that can start some games, platoon a little, and have some value. But you better look for an improvement if you really want to contend, because a .330/.330 guy really isn’t going to cut it, unless his fielding and base running is spectacular.

The “if” is an issue, though, as these projections aren’t set in stone. Perhaps Gwynn has been rejuvenated by the move to San Diego. Maybe, just maybe, he’s elevated his play due to the high expectations. It’s always tough to detect if there’s truly been a change in talent level, or if unexpected performance is just random fluctuation in the numbers (a hot performance over a few weeks). As numbers geeks, it is easy for us to take the safe route, the often time correct route, and say that Gwynn is who we thought he was – a .600/.700 OPS guy. But if he gets sent somewhere else, and turns into a perennial all-star, that wouldn’t sit well with the San Diego faithful.

It’s up to the Padres, with their scouting, their stats, their video, to determine who Tony Gwynn Jr. is going forward. For now, we can just sit back and watch, and hope that this Hollywood story continues to write itself.

Stolen base research

Along with misinterpreting a stat in my last post, I also somehow managed to wipe out another short post I had made on the same topic. Good grief! I guess this truly means I’m back ; )

Anyway, before looking into the Pads’ 2009 base running stats, I discovered a few recent base stealing articles on the ‘net. One, Brian Cartwright’s latest in the BP Idol contest, is kind of a running (retrosheet) history of the stolen base. Gives a couple nice charts at the end that show how stolen base percentage has been on the rise, while stolen base attempts have been in decline.

The second, by Dan Turkenkopf at THT, gives us a couple of alternative ways to measure the value added from base stealing (including Net Stolen Bases, developed by Rich Lederer and slightly altered by Dan, which I used in my last post).

Good stuff.

2009 Padres base running

What follows is simply for entertainment purposes. The dreaded sample size issue renders most of it useless; still, I think it is at least somewhat insightful, if only to point out what has happened, not what will happen.

Anyway, Baseball Reference has some great base running stats. I messed around with some stuff, and took out what I thought was most useful: Here are two charts with explanations after each one:

Player PA RS% SB CS PO NET SBA%
Team 2593 8% 27 12 10 -7 4.4%
Gonzalez 297 9% 1 1 1 -2 1.7%
Eckstein 273 10% 2 1 0 0 2.5%
Kouzmanoff 269 4% 1 0 0 1 1.1%
Giles 254 6% 1 0 0 1 1.3%
Headley 239 5% 6 2 0 2 12%
Hundley 163 4% 1 1 2 -3 .3%
Hairston 161 9% 6 1 1 3 14%
Rodriguez 122 9% 0 0 0 0 0%
Gerut 121 11% 2 0 0 2 5.6%
Gwynn 107 16% 2 3 3 -7 8.2%
EGonzalez 104 8% 0 2 0 -4 8%
Blanco 102 8% 0 0 0 0 0%

 

PA – Plate appearances – simply used as a cutoff point (100 PAs)
RS% – Percentage of time player scored once they reached base — correction: turns out RS% isn’t what I thought it was. That kind of screws up my thoughts on Gwynn below, too. It is actually: (R-HR)/(AB+H+BB-HR). Thanks to Peter in the comments for making me aware of this.
SB – Stolen bases
CS – Caught stealing
PO – Pick offs
NET – SB-(CS*2)-PO
SBA% – Percentage of times player attempts to steal when they have the opportunity

continued ….

Player OOB XBH% 1stS R3rd R3rd% 1stD RH RH% 2ndS RH RH%
Team 23 37% 112 29 26% 33 16 49% 56 29 52%
Gonzalez 4 28% 20 2 10% 5 3 60% 7 4 57%
Eckstein 2 38% 14 4 29% 4 1 25% 11 6 55%
Kozumanoff 3 38% 9 2 22% 4 2 50% 3 2 67%
Giles 3 44% 11 5 45% 1 0 0% 4 2 50%
Headley 1 29% 8 3 38% 2 0 0% 4 1 25%
Hundley 1 33% 4 0 0% 4 2 50% 4 2 50%
Hairston 1 63% 4 3 75% 2 2 100% 2 0 0%
Rodriguez 2 50% 5 1 20% 0 0 - 3 3 100%
Gerut 1 38% 5 1 20% 3 2 67% 0 0 -
Gwynn 2 25% 12 2 17% 2 1 50% 2 1 50%
EGonzalez 0 88% 5 4 80% 1 1 100% 2 2 100%
Blanco 1 17% 4 0 0% 1 1 100% 1 0 0%

 

OOB – Outs on base, not including caught stealing, pick offs, or force outs
XBH% – Percentage of times the runner advances more than one base on a single or two on a double, when possible (as defined by B-R)
1stS – Number of times player is on first when single is hit
R3rd – Number of times player reaches third when single is hit
R3rd% – R3rd divided by 1stS
1stD – Number of times player is on first when double it hit
RH – Number of times player reaches home when double is hit
RH% – RH/1stD
2ndS – Number of times player is on second when single is hit
RH – Number of time player reaches home when single is hit
RH% – RH/2ndS

****

Okay, so we’ve got a lot of numbers there. In the end, like I mentioned, I’m not sure there is really much we can glean from them. The opportunities in most cases are so scarce. For real analysis, we’d have to do this over multiple seasons, not half of one.

Anyway, the player that sticks out is Gwynn. He scores 16% of the time he reaches base (wrong: see correction above), which is clearly highest on the team. His stolen base numbers, however, have been bad. He’s been caught three times in five attempts. He’s also been picked off three times. Gwynn has attempted a steal in about 8% of his opportunities, good for third highest on the team among qualifiers.

Gwynn has also been bad at advancing on hits, as he’s taken the extra base 25% of the time, which is only ahead of Henry Blanco on the Padres. So, Gwynn has been bad at stealing bases and he’s been bad at advancing on hits … yet, he’s scored a lot when he gets on base.

Go figure. Seriously, that exercise there only magnifies was I was saying earlier. These numbers, for predicting anything, are probably close to useless. With so few opportunities, context can really cloud the numbers. Say, for instance, that Gwynn has found himself of first with a rocket hit right at the left fielder a few times, or if the ball’s hit to a guy with a great arm – there is little he can do there do make it to third, yet those things aren’t reflected in these relatively simple numbers. Over the long haul, things like that tend to even out, and that’s part of the reason why the numbers gain significance as we look at more data.

Second base fielding, THT style

We’re moving on to second base in an early glimpse at 2009 fielding performance, according to stats available at The Hardball Times. Check this link for a little background on what we’re doing. Here are the second base top/bottom 5, then the Padres:

Top 5

Player Team RAA
Hill TOR +15.2
Phillips CIN +10.1
Weeks MIL +9.2
Kendrick LAA +8.8
Zobrist TB +6.9

 

Bottom 5

Player Team RAA
Schumaker STL -9.5
Uggla FLA -9.2
Kennedy OAK -8.6
Getz CHA -8.2
Collaspo KC -6.8

 

Padres

Player Innings RAA
Eckstein 519 -4.1
Gonzalez 73 0

 

Eckstein hasn’t been very good, according to these numbers. UZR has him at –.5 runs, so it’s clearly too early to tell if he’s really lost a step or not. Over the past two seasons, however, UZR has Eckstein at about –15 runs total (at shortstop). For his career, he’s about 2 runs below average per 150 games.

For the Blue Jays, Aaron Hill appears to be a star defensively at second. According to UZR, he’s +10 per 150 for his career. Though, interestingly, he’s –8 runs at short in about a half a season’s worth of innings (mostly in 2006).

In last place is Skip Shumaker who has been bad by UZR, as well, at –9 runs. His double play partner, Brendan Ryan, leads shortstops with a +12.7 mark (THT stats) so far.

Quick hits: Draft recaps

Geoff Young has some thoughts on the Padres’ draft, among other things, in his must-read weekly post.

Peter Friberg interviews prospect expert Johns Sickels about the draft. Sickels is pretty optimistic about the Padres’ selections, and he thinks all three early high school picks (Tate, Williams, Sampson) will sign.

At Mad Friars, John Conniff interviews Kevin Goldstein and Jim Callis. Both guys are very excited about the Pads’ picks.

Peavy’s injury

Daniel links to Will Carroll’s take on Peavy’s injury. Carroll thinks Peavy is essentially done for the year, although that assessment may be, as Will calls it, a bit provocative.

Still, if Carroll is right about the injury, a return in 3-4 weeks is out of question, and the injury is definitely more serious than the initial prognosis. Hopefully, the Padres will be extra-safe with Jake, considering the circumstances around this season (I don’t expect them to seriously compete down the stretch).

Here’s Peavy’s take on things.

****

I’m having some computer issues, so things may be a bit slow for a while. I’m hoping I’ll get it fixed soon, but we’ll see. Can you tell I don’t have great luck with computers?

What’s wrong with Kouzmanoff?

Nothing. Maybe.

I was going to investigate what’s been wrong with Kouz this year. I started, actually. Then, after perusing his Fangraphs card, I stopped. Sure, there are a couple of angles I could take (his production is worse against the fastball this year, for instance), but after thinking about it for a second, there is little I can add from a numbers-perspective.

The main problem is the usual one, sample size. I understand that it is a boring take, and sometimes it looks lazy (maybe it is), but it is really the most honest approach. ZiPS’ projection for Kouz, for the rest of the season, is .260/.311/.441 (.328 wOBA). That’s decent production, somewhere right between Kouz’s ’07 and ’08 performance. His fielding, according to UZR, is right around average (where it’s been for his career).

Kouzmanoff is in a slump offensively, no doubt. It is a relatively long slump. His projected future performance is now a shade below where it was when the season started. That is about as much as we can say.

Sure, we could look at his batted ball distribution and say he’s a different hitter (a worse one) or we could look at another number and say something else. We could paint a story however we’d like, and use some numbers to back it up.

In the end, however, I think we’re better off sticking by our initial projections, and making some minor adjustments along the way. A player can change a lot in a year, but it’s generally not going to show up in a few hundred plate appearances. In other words, it’s tough to tell if it’s a true change in performance level, or merely a slump. That is why it’s better to rely on the long term projections, which incorporate years worth of data.

If we were to incorporate scouting, perhaps we could make more out of early season performance. Maybe the new technological advances (like PITCH and HIT f/x) will help bridge the gap in the coming years. But for now, if we’re going to analyze things strictly based on numbers, we have to realize our limitations.

Follow

Get every new post delivered to your Inbox.