Betancourt and Batista, You Both Suck

by Graham on Aug 27, 2008 7:40 PM PDT Comment 59 comments

Combined, you are on the order of 6 wins worse than average over a full season. (Assuming Yuni's a -10 defensive shortstop, which is being generous).

In bRAA (wOBA-based) and pRAA (tRA based), your Seattle Mariners:

Yes, I know. But Kenji doesn't completely suck at defence.

Marinerpitching08

Bahahahahahaha how is this possible.

0 recs | Comment 59 comments

Story-email Email Printer Print

Comments

Poor Beltre

by Fogel on Aug 27, 2008 7:45 PM PDT reply actions 0 recs

This is what inspires me to be an alcoholic

I fucking hate you Mariners

by kentroyals5 on Aug 27, 2008 7:47 PM PDT reply actions 0 recs

And Kenji's line manages to be worse.

by Two Rs and Two Ls on Aug 27, 2008 7:49 PM PDT reply actions 0 recs

Ahem!

Yes, I know. But Kenji doesn’t completely suck at defence.

by Graham on Aug 27, 2008 7:50 PM PDT up reply actions 0 recs

Granted, but still impressive.

by Two Rs and Two Ls on Aug 27, 2008 7:52 PM PDT up reply actions 0 recs

Top three hitters:

Raul, Greg Norton, Felix Hernandez.

Your 2008 Seattle Mariners.

by justcougit on Aug 27, 2008 7:50 PM PDT reply actions 0 recs

Lopez was just at +4 about a week ago :(

by Fogel on Aug 27, 2008 7:53 PM PDT up reply actions 0 recs

*probably more than a week

by Fogel on Aug 27, 2008 7:53 PM PDT up reply actions 0 recs

the wOBA's were corrected a few days ago.

I wasn’t counting RBOE amongst the league totals due to a SQL issue. So 2007 and 2008 individual values went universally down.

by Matthew on Aug 27, 2008 9:22 PM PDT up reply actions 0 recs

:(

by Fogel on Aug 27, 2008 9:25 PM PDT up reply actions 0 recs

That's why we call it beta!

by Graham on Aug 27, 2008 9:39 PM PDT up reply actions 0 recs

Felix, in one at bat...

Twice as valuable at the plate as Ichiro all year.
That blows my mind.

by AnotherAaron on Aug 27, 2008 9:50 PM PDT up reply actions 0 recs

It would likely go way down if he had more AB's

The small sample size skews it.

by OlSalty on Aug 27, 2008 10:25 PM PDT up reply actions 0 recs

Not really. A season of average >> an at-bat of awesome.

by Graham on Aug 27, 2008 10:45 PM PDT up reply actions 0 recs

Also

You can find this sort of thing for each team at StatCorner. Which I would totally be pimping even if I didn’t help run it.

by Graham on Aug 27, 2008 7:51 PM PDT reply actions 0 recs

Is there a reason you have O-xO but not R-xR?

by Fogel on Aug 27, 2008 7:54 PM PDT up reply actions 0 recs

We don't have R yet.

Requires doing some extra work that we haven’t got around to (and is why we don’t have SB/CS either)

by Graham on Aug 27, 2008 8:04 PM PDT up reply actions 0 recs

Pimp it.

You are an admin and it’s not like you would be spamming.

by Wilder. on Aug 27, 2008 8:40 PM PDT up reply actions 0 recs

On one hand, our rotation has been worse than theirs, but on the other hand, Jesus at least we have some green pitchers

by Fogel on Aug 27, 2008 7:52 PM PDT reply actions 0 recs

None of their guys are particularly far below average

thus, they pwn us

by seattlebruin on Aug 27, 2008 8:42 PM PDT up reply actions 0 recs

How much of an adjustment do SS get?

I’m guessing average SS is probably -5 to -10 offensively so Yuni isn’t quite so bad. This makes him around -20 compared to average SS out there. He still sucks but Vidro is worse because an average DH is probably 10 runs above average so Vidro was around -25 runs and that was in less playing time.

by Edgar for Pres on Aug 27, 2008 8:04 PM PDT reply actions 0 recs

DH is -5, SS is +10 IIRC

And defence matters.

by Graham on Aug 27, 2008 8:10 PM PDT up reply actions 0 recs

Almost

The position adjustment levels, as developed by Tango:

Catcher: +15 runs
SS/CF: +5 runs
2B/3B: No adjustment
LF/RF: -5 runs
1B: -10 runs
DH: -15 runs

The gap between an average SS and an average DH is about 20 runs. So, yes, Vidro is still significantly worse.

by davidcameron on Aug 27, 2008 8:23 PM PDT up reply actions 0 recs

How bad is Yuni's D?

He’d be within half a win if it were even -10. Although not by rate, obviously.

by Graham on Aug 27, 2008 8:25 PM PDT up reply actions 0 recs

It's bad

Fielding Bible +/- has him at -21 plays so far this year, which would be about -15 runs. They had him at -10 plays last year and -3 in 2006, so Dewan’s system sees the same ridiculous collapse of defensive value we’ve all noticed.

So, yea, you can argue that Vidro and Betancourt have hurt the M’s about equally this year, though it took Vidro about half the playing time, which makes his remarkable suckitidue more impressive (depressive?)

by davidcameron on Aug 27, 2008 8:28 PM PDT up reply actions 0 recs

Holy crap Corcoran has been kick-ass

by OlSalty on Aug 27, 2008 8:15 PM PDT reply actions 0 recs

8.3% LD%

He and Carlos Marmol have basically identical line drive rates, and they’re lapping the field (among pitchers with at least 50 IP). The next best guy after thsoe two is Scott Downs at 12.3%.

by davidcameron on Aug 27, 2008 8:31 PM PDT up reply actions 0 recs

one question

I"m kind of new at this, so I have a question. When looking at bRAA and pRAA for the entire league, most players are at negative values, and there are a few people like Pujols and Cliff Lee with such high values that they seem to be skewing the average upwards. So when we look at these numbers, is 0 still what we are considering average despite possible positive skews? And would batting/pitching runs above median provide better information regarding who has been playing better or worse than most?

by boomdonkey on Aug 27, 2008 8:37 PM PDT reply actions 0 recs

No - average is in relation to the whole, not the number of players that are better or worse

an “average” player is never an actual player on some team, but a concept of what would happen if you took all the players in the league at one position and lumped them into one, dividing by 14. Essentially.

by seattlebruin on Aug 27, 2008 8:45 PM PDT up reply actions 0 recs

Graham/Matthew

What is the median bRAA and pRAA? A distribution would be even better. A PA/IP normalized distribution would be the best. If this takes more that a few minutes though then don’t bother.

by Edgar for Pres on Aug 27, 2008 8:56 PM PDT up reply actions 0 recs

It's something we'll put up eventually

But it’s not to hand, so will have to wait.

by Graham on Aug 27, 2008 8:58 PM PDT up reply actions 0 recs

`

For all batters: -1.20
For PA > 100: -2.25
For all pitchers: -0.36
For xIP > 15: -0.24

by Matthew on Aug 27, 2008 9:03 PM PDT up reply actions 0 recs

FWIW, there are not surprsing values

players that suck don’t get the chance to rack up huge negative RAA totals like players that excel are able to do with huge positive RAA totals. So yes, there’s skew but I’m pretty sure the overall talent level in the major leagues is not a normal distribution.

by Matthew on Aug 27, 2008 9:06 PM PDT up reply actions 0 recs

Incidentally Tango has looked at this a little bit

http://www.tangotiger.net/talent.html

by Jeff on Aug 27, 2008 9:11 PM PDT up reply actions 0 recs

Yeah, one of the interesting things myself and the Wharton group found when doing research for SAFE

was that on an individual talent basis (things like arm strength, power, velocity), MLB almost always had close to a normal distribution. It’s just when you put it all together than you introduce the tail distribution (with playing time moving it back toward normal)

by Matthew on Aug 27, 2008 9:20 PM PDT up reply actions 0 recs

Thanks for the info everyone

by boomdonkey on Aug 27, 2008 9:22 PM PDT up reply actions 0 recs

Thats pretty close to zero

by Edgar for Pres on Aug 27, 2008 9:15 PM PDT up reply actions 0 recs

I also have a question

Isn’t wOBA, much like prOPS, rooted in a hitter’s balls in play? Shouldn’t Beltre look much better then in terms of bRAA?

by CKel on Aug 27, 2008 8:45 PM PDT reply actions 0 recs

No.

wOBA = (0.72xNIBB + 0.75xHBP + 0.90×1B + 0.92xRBOE + 1.24×2B + 1.56×3B + 1.95xHR) / PA

by Matthew on Aug 27, 2008 8:46 PM PDT up reply actions 0 recs

Would it be not worth it to apply the prOPS formula to wOBA?

by Fogel on Aug 27, 2008 9:05 PM PDT up reply actions 0 recs

PrOPS has a good goal but poor execution.

It’s discussed in the “A Common Mistake” thread.

by Matthew on Aug 27, 2008 9:07 PM PDT up reply actions 0 recs

Right but wouldn't it be better than nothing?

Maybe?

by Fogel on Aug 27, 2008 9:12 PM PDT up reply actions 0 recs

It'd be taking a great stat and half-assing a regression for luck.

I (and maybe others if I can convince them to help me out) will be working on a robust way of actually doing this in the future.

by Matthew on Aug 27, 2008 9:15 PM PDT up reply actions 0 recs

Suggestion

by Fogel on Aug 27, 2008 9:16 PM PDT up reply actions 0 recs

Umm...

by Matthew on Aug 27, 2008 9:18 PM PDT up reply actions 0 recs

Attempt

if you apply Beltre’s career LD/GB/FB/etc. outcomes to his 2008 performance, you come out with a wOBA of .327.

This, of course, makes him come out worse since he’s been in the Majors from a young age at which he hadn’t yet fully developed his power. So in reality he’s been better than that figure would suggest.

by Jeff on Aug 27, 2008 9:23 PM PDT up reply actions 0 recs

Ah so the outcomes as opposed to the type of hit

Thanks

by CKel on Aug 27, 2008 8:50 PM PDT reply actions 0 recs

You may have explained this and I'm not seeing it.

But Fangraphs has Betancourt at -29.74 (-7 worse than Johjima). Is there a reason for the discrepancy?

...and now I'm here

by Librocrat on Aug 28, 2008 3:34 AM PDT reply actions 0 recs

Fangraphs' BRAA is based on run expectancy.

Wheras BRAA based on wOBA is not. So Fangraphs takes into account how many runners are on base and how many outs there are. wOBP does not.

There may be more to it than that, but there’s one difference.

by Teej on Aug 28, 2008 4:16 AM PDT up reply actions 0 recs

*whereas

by Teej on Aug 28, 2008 4:18 AM PDT up reply actions 0 recs

So is Yuni basically the model of Replacement Level Player?

Or is he even worse?

by Jed MC on Aug 28, 2008 8:09 AM PDT reply actions 0 recs

Depends on his defense.

I’d think he’s worse than replacement.

by Matthew on Aug 28, 2008 9:34 AM PDT up reply actions 0 recs

I'm having trouble reconcilliing that, judging from wOBA*,

Angels stadium is somehow simultaneously assisting our hitters AND pitchers at the same time. This seems logically impossible.

~Till the Halo burns out...

by Zu Long on Aug 29, 2008 9:37 AM PDT reply actions 0 recs

That's really, really weird

Angels stadium apparently inflates singles and doubles while repressing line drives, and since the first two go into wOBA and the last one goes into tRA…

by Graham on Aug 29, 2008 10:19 PM PDT up reply actions 0 recs

Yeah, I figured as much. And more than likely it's the singles that really kill us.

107% is more than anyplace else in the game IIRC. There are so MANY singles in a give season that even after it’s halved, that’s a huge boost to our expected runs. I’m honestly not sure how you adjust for that.

~Till the Halo burns out...

by Zu Long on Aug 31, 2008 2:52 PM PDT up reply actions 0 recs