What happened to Yuniesky Betancourt?

[rluzinski]

Ray Durham was a freqeunt 3 to 4 WAR player for a long stretch of his career. Betancourt has never been above 2.2. I'm sorry, but that's a terrible comparison.

 

As is using WAR to decide how good a player is.

Using 4+ years worth of fangraph's WAR is a pretty decent way of estimate their true talent over that period. I don't think there is a better quantitative way of doing it right now. I wouldn't use one year's worth of war without regressing it a lot, tough.

[Thurston Fluff]

I've said everything I've needed to say, and presenting more stats isn't going to sway anyone anyways. There's a mountain of data that's out there, and if at this point, people want to ignore it, that's their prerogative.

 

Then use those stats instead of always going back to WAR like it's the end all be all of stats that cannot be argued. I agree Yuni is bad but WAR is just a bad stat.

[DuWayne Steurer]

I've said everything I've needed to say, and presenting more stats isn't going to sway anyone anyways. There's a mountain of data that's out there, and if at this point, people want to ignore it, that's their prerogative.

 

Then use those stats instead of always going back to WAR like it's the end all be all of stats that cannot be argued. I agree Yuni is bad but WAR is just a bad stat.

WAR is quick and easy, and like Russ said, 4 + years worth is pretty telling. I'm not going to argue the merits of WAR in this thread though.

[RockCoCougars]

Again, does anybody know why Escobar has such a high offensive WAR on baseball reference, is it league based or something? And as for Greinke, I never panicked. As I've said, I was just glad to dump stiffs like Escobar (who was in my view one of the worst regular players in Brewers history in 2010) and Jeffress. Cain will probably be a good not great player and Odorizzi is a coin flip (probably worse had he stayed considering the Brewers historical luck with young pitchers).

[Thurston Fluff]

WAR is quick and easy, and like Russ said, 4 + years worth is pretty telling. I'm not going to argue the merits of WAR in this thread though.

 

If you are going to use it then you should be prepared to argue it's merits. Quick and easy is fine but good and accurate are better.

 

Using 4+ years worth of fangraph's WAR is a pretty decent way of estimate their true talent over that period.

 

I think using four years as a composite is ok but not compiling four single seasons into one average. My main reason for that is the same issues with single season UZR accuracy is there as opposed to four years of stats rolled into one. A four year WAR instead of averaging four single season WARS together would eliminate that pesky problem of single season UZR inaccuracy that renders single season WAR useless. Merely averaging single season ones together doesn't eliminate that problem.

[topper09er]

Again, does anybody know why Escobar has such a high offensive WAR on baseball reference, is it league based or something?

At baseball-ref offensive WAR = Runs above average from batting, base running, reaching on errors, avoiding DPs, positional adjustments and replacement level.

Escobar:

Year	Age	Tm	Lg	PA	Rbat	Rbaser	Rroe	Rdp	Rfield	Rpos	Rrep	RAR	WAR	oRAR	oWAR	dWAR
2011	24	KCR	AL	437	-17	1	1	2	11	6	15	18	1.8	7	0.7	1.1

So just from batting he is 17 runs below average, but when you add everything together for offense you get to 7 runs above replacement, for 0.7 oWAR.

 

Yuni looks like this:

Year	Age	Tm	Lg	PA	Rbat	Rbaser	Rroe	Rdp	Rfield	Rpos	Rrep	RAR	WAR	oRAR	oWAR	dWAR
2011	29	MIL	NL	405	-10	1	-2	-1	-4	5	11	-0	0.0	4	0.4	-0.4

Yuni is at 10 runs below average from batting, and comes out to 4 runs above replacement on offense, for 0.4 WAR

 

The main differences is that Escobar gets 15 runs to Yunis 11 runs from replacement level because the AL is a stronger league so league average in the AL means you are further from replacement level than in the NL. The average NLer is worse than the average ALer. Escobar also has a little more playing time which helps because WAR is also a counting stat dependent on playing time. This is why Escobar get 6 positional runs to Yuni's 5.

[RockCoCougars]

Again, does anybody know why Escobar has such a high offensive WAR on baseball reference, is it league based or something?

At baseball-ref offensive WAR = Runs above average from batting, base running, reaching on errors, avoiding DPs, positional adjustments and replacement level.

Escobar:

Year	Age	Tm	Lg	PA	Rbat	Rbaser	Rroe	Rdp	Rfield	Rpos	Rrep	RAR	WAR	oRAR	oWAR	dWAR
2011	24	KCR	AL	437	-17	1	1	2	11	6	15	18	1.8	7	0.7	1.1

So just from batting he is 17 runs below average, but when you add everything together for offense you get to 7 runs above replacement, for 0.7 oWAR.

 

Yuni looks like this:

Year	Age	Tm	Lg	PA	Rbat	Rbaser	Rroe	Rdp	Rfield	Rpos	Rrep	RAR	WAR	oRAR	oWAR	dWAR
2011	29	MIL	NL	405	-10	1	-2	-1	-4	5	11	-0	0.0	4	0.4	-0.4

Yuni is at 10 runs below average from batting, and comes out to 4 runs above replacement on offense, for 0.4 WAR

 

The main differences is that Escobar gets 15 runs to Yunis 11 runs from replacement level because the AL is a stronger league so league average in the AL means you are further from replacement level than in the NL. The average NLer is worse than the average ALer. Escobar also has a little more playing time which helps because WAR is also a counting stat dependent on playing time. This is why Escobar get 6 positional runs to Yuni's 5.

Thank you for the reply. I figured that it had to be partially league based or something. Also didn't know the different baseball sites calculated WAR differently.

[rluzinski]

I think using four years as a composite is ok but not compiling four single seasons into one average. My main reason for that is the same issues with single season UZR accuracy is there as opposed to four years of stats rolled into one. A four year WAR instead of averaging four single season WARS together would eliminate that pesky problem of single season UZR inaccuracy that renders single season WAR useless. Merely averaging single season ones together doesn't eliminate that problem.

WAR is a counting stat, so you can just add it (and then regress if you are looking for an estimate of true talent). Overall playing time does come into play, though. Just like comparing HR totals, a guy who's played 600 games over 4 years is going to have higher total than one who played 300, all else equal. Of course, if one guy played more because he was more durable, I don't think you automatically ignore that.

UZR/150 is a rate stat, so you have to be careful about just averaging them if the sample sizes are different. For instance:

Player 1: 5 UZR/150 (150 games)

Player 2: 5 UZR/150 (30 games)

The first is "better" in two ways. First, he saved more overall runs (5 runs vs player 2's 1 run). Second, it is more likely that player 1 is the superior defender (although by no means is it certain).

"Thank you for the reply. I figured that it had to be partially league based or something. Also didn't know the different baseball sites calculated WAR differently."

WAR is just a basic framework. I think it is a good thing that different sites are using slightly different methodologies when utilizing that framework. A "living" metric will be allowed to get better as time goes on.

[mothershipconnection]

kramnoj, Betancourt is the poster child for the "we're smarter than you are crowd because we use fancy sabermetric stats that weren't around 20 years ago" and only we can interpret them for you out of date traditionalists that actually watch guys play and decide for yourself

 

I'd like to know who said they're smarter than you because they use numbers. Otherwise, ease off the martyrdom a little bit, especially the "unlike you, I watch games" tactic. Are you suggesting that those who disagree with you don't actually watch baseball games?

[Ennder]

Betancourt has no range and his approach at the plate is just awful, not like watching the games makes him look good either. I hope we find someone better next year, I don't know that we will though.

[brettac1]

I can't say I have noticed any difference with his "approach", no matter what Bill Schroeder and Sveum tell me. Same awful stuff as always.

[ECpackers]

....he wasn't hitting. and now he is hitting. that is not awful in the least.

[JJHardy7]

Well, I'm out, we've got 2 Betancourt threads where we're arguing whether he's just REALLY bad, or REALLY REALLY bad.

 

I guess I'm personally sick of the how really, really bad stuff with Yuni. I think a fair compromise would be to say he's really, really helped the offense the past month. Doesn't mean he's great...but the guy in some views here will stink no matter what. He is what he is...I'd just like some to say he's had a good month offensively. That is all. Just admit that without taking a shot saying but this, that, etc. It's gotten old. That's why the last thread and this thread will go on and on and on. Yuni isn't the worst baseball player in the history of baseball players. He has some big weaknesses...but I think a major part of why we're sitting where we are sitting over the past month...is Yuni B's offense. He could start stinking again...but I'll give him props when due. And not add...but he still stinks! Let's enjoy the ride.

[brettac1]

Still has that sweet leather, I say.

[pitchleague]

I'm just hoping that Betancourt will pull an Edgar Renteria in late October and then, just for a few minutes, EVERYBODY on this site will be happy that we have him on our team. If only for a moment, of course.

 

A guy can dream . . . .

[Ennder]

....he wasn't hitting. and now he is hitting. that is not awful in the least.

He has one of the worst approaches at the plate that I've ever seen. When his balls are finding holes it looks ok, when they aren't his bat is mostly worthless. That is how most mediocre low OBP players are, he isn't unique at all. When he is hot he looks ok, the rest of the time he is a drain on the team.

[PKBadger]

If I can find a picture of Yuni doing that "grizzly bear roar" thing the team has been doing when they get hits, I'm making it my avatar.

[Team Canada]

Maybe it's working for him right now, but Yuni's one-handed swing is incredibly ugly to me.

[Thurston Fluff]

WAR is a counting stat, so you can just add it (and then regress if you are looking for an estimate of true talent).

 

I think we can use WAR effectively in this instance by combining all the numbers into one number not by adding then averaging them. The main reason is the same as always. UZR is not a stat that gives any useful information in one year quantities. To simply take 4 misused calculations add them and average them is to use four bad pieces of information and expecting them to give you something good. That can be solved by adding all the data from four years into one WAR.

 

For instance if one season's UZR is off because he played really well/poor for a 2 month stretch (or got un/lucky) that year's UZR would be significantly effected by it because that is 30% of the season. To then average that WAR with 3 other years WAR would mean the odd one counts for 25% of the average. Now take that same 2 month stretch and add it into all the other months over a 4 year stretch and it's effect is much less than it would have been by adding and averaging the four single years.

[rluzinski]

For instance if one season's UZR is off because he played really well/poor for a 2 month stretch (or got un/lucky) that year's UZR would be significantly effected by it because that is 30% of the season. To then average that WAR with 3 other years WAR would mean the odd one counts for 25% of the average. Now take that same 2 month stretch and add it into all the other months over a 4 year stretch and it's effect is much less than it would have been by adding and averaging the four single years.

The issue isn't whether you are summing or averaging. In your example, the odd 2 months represents 30% of 25%(7.5%) of the total either way. The average is just the total divided by 4, after all.

The issue of estimating true talent with just WAR is that each component of WAR (offense, fielding and base running) needs to be regressed by different amounts. The correct way (I think) is to take the components of WAR and regress them each separately before adding them up. The better a particular metric can capture true talent, the less you have to regress.

The interesting thing is that if you just care abotu what the player did (you aren't estimating true talent), you still need to regress UZR. You don't need to regress the offensive stats, since a single is a single. Here is an interesting thread on the topic:

http://www.insidethebook.com/ee/index.php/site/comments/does_one_year_uzr_tell_us_exactly_what_happened/

They also discuss evidence of a systematic bias in the raw data used by UZR.

[1992casey]

I just thought I'd mention that there's a significant edit in the message directly above, just in case anyone read this topic in the hour between the initial posting and the edit.

[topper09er]

With Betancourt, you dont even need UZR, just watch the games. He is horrible out there, just last night he tried his hardest to lose the game, not even because he is fat and slow, but because he is dumb and lazy. Who tries to tag out a stealing runner on a grounder? Why would it even occur to him to try and do that, he has to be thinking with 2 outs and the runner going if he gets his glove on it throw to first. Instead he just tries to take the easy way out and tag the closest guy and drops the ball....and doesnt even hustle to pick it up and still maybe get the runner at first. We cannot have him out there in the playoffs, a good team will kill you if you give them 4 outs 6 times in 5 games. He is going to do something like that in the playoffs and will become the Bill Buckner of Milwaukee

[The Truth]

yep. Betancourt almost cost the Brewers the game last night. Especially with Pujols due up and a RISP situation. We got lucky there. Dodged a bullet. Brewerfan.net would have exploded if Pujols had gotten a walk off hit there instead of popping up

[reillymcshane]

With Betancourt, you dont even need UZR, just watch the games. He is horrible out there, just last night he tried his hardest to lose the game, not even because he is fat and slow, but because he is dumb and lazy.

This may be a little harsh in the wording, but it sums Yuni's fielding for me. Some people disagree, no doubt, but I think that just watching him, he is terrible in the field. UZR only reinforces how bad he is.

[RockCoCougars]

Everyone makes errors like that at some point. I'll give you that it was a bonehead play, but it was kind of an awkward play with the ball and the runner coming at the same time right near the bag. I can't think of one game that Betancourt has directly cost the team this season due to poor fielding. It seems that there may have been one game where he made an error and the opponent scored about 5 runs with the extra out, but that's all that comes to mind. Last year, Escobar made a handful of crucial errors that directly led to a loss.