We all Know Braun's Having a Great Year, but what about his OBP?

[kramnoj]

http://jinaz-reds.blogspot.com/2008/06/why-do-i-keep-using-ops.html is the link. The difference is small, but it was revealing to me.

[logan82]

Also note that wOBA, which roughly equates to (OBP*2+SLG)/3, has an even higher correlation to runs scored.

[TooLiveBrew]

yep, and GPA, too. Though it's only fair (& highly relevant) to include this part:

 

"Big improvement by going from AVG to either OBP or SLG. Pretty big jumps from SLG on up to OPS. But not much of an improvement after you hit OPS. The most accurate stats proved to be those based on linear weights: RAA/PA, wOBA, and R/G. That's gratifying, as I've been treating them as the gold standard statistics on this site for rating hitters... but they're incremental improvements, at best, over plain old OPS in terms of predicting runs scored."

 

From the graph displayed, basically everything from OPS to R/G is about the same. Not exactly, but close enough for this simple Simon.

 

Basically, BA < OBP/SLG < OPS/etc. -- from this study, anyway. Great find, and thanks for sharing that kramnoj

[Brawndo the Thirst Mutilat]

Here's Tango, using linear weights run values (which the guy from the Reds blog correctly identified as the "gold standard" in run estimation), to demonstrate mathematically that 1.7 x OBP + SLG is correct:

http://www.insidethebook.com/ee/index.php/site/comments/why_does_17obpslg_make_sense/

This is a quote from Dan Szymborski, inventor of ZiPs, regarding the noted higher Pearson r value for team SLG relative to team OBP in the previously linked study:

On the team level, that wouldn't be surprising - team OBP tends to be clustered together a little more than team SLG. Karl Pearson wasn't a demi-god - r isn't the sole determinant of utility.

I'd link to the BTF discussion thread where this was said, but the link is so long it breaks the page width (good thing I previewed). Go to BTF and read the comments for the "don't press the red button" article from yesterday.http://www.baseballthinkfactory.org/files/newsstand/discussion/bugs_cranks_kyte_whatever_you_do_dont_press_the_big_red_button/
Because there is always a smaller spread between team OBP, team SLG tends to have higher r-values relative to runs scored. This is true not just for recent years, but for vast chunks of baseball history. Post 2 here looked at 40 years of data:

http://www.baseball-fever.com/showthread.php?t=48531

r values: OBP = .904, SLG = .926.

This does NOT mean that 1 marginal point of SLG > 1 marginal point of OBP, nor that SLG is more important than OBP. In fact, 1.8 x OBP + SLG had a higher r value than simple OPS in this study. There's no real consensus about whether 1.7 or 1.8 is the correct number to use, but that's really the only thing up for debate anymore (i.e. not whether OBP or SLG is more important, just the factor by which OBP is more important).

[kramnoj]

values: OBP = .904, SLG = .926.

 

This does NOT mean that 1 marginal point of SLG > 1 marginal point of OBP, nor that SLG is more important than OBP.

 

I'm FoP in the where Szymborsk has that remark, but it's not clear to me what that quote means. I'm far from an expert in stats, but when I hear that slugging correlates higher to runs, I understand that having a really high slugging is more important than a really high OBP (not by much, granted, but that's what the stats seem to say). Now, I don't understand why you felt the need to capitalize the NOT, there, nobody was claiming that one point of slug is better than OBP. What I don't understand is why slug isn't more important than OBP if it correlates higher on a team level to runs.

 

If I just try to understand this in the common sense of observation, I would say that for the vast majority of players, the range of OBP is from .300 - .400 (the right end isn't exact, just a round number), whereas for slugging it again starts at around the .300 level but extends to .600 (maybe .575). Because the range for OBP is smaller, a marginal gain in OBP is worth more than an equal gain in slugging. That makes sense to me. But I think I would prefer a guy who is leading the league in slugging than a guy who leads the league in OBP.

[Postseason2006]

Can we also find out what his Wonderlic and IQ is too?

[iluvlamp]

I just read this thread. It's obvious that Braun will have to conform to a set skill set. He is just not good enough. I have my own Braun equation.

 

Ryan Braun > Ryan Adams > Brian Adams > Brad Ausmus > Bradley Univeristy > Unitards > OBP

[Brawndo the Thirst Mutilat]

I'm FoP in the where Szymborsk has that remark, but it's not clear to me what that quote means. I'm far from an expert in stats, but when I hear that slugging correlates higher to runs, I understand that having a really high slugging is more important than a really high OBP (not by much, granted, but that's what the stats seem to say). Now, I don't understand why you felt the need to capitalize the NOT, there, nobody was claiming that one point of slug is better than OBP. What I don't understand is why slug isn't more important than OBP if it correlates higher on a team level to runs.

 

I wondered whether that might be you. We're in much the same boat in terms of statistical expertise - it's a lot like another language, and I'll confess I'm better at understanding it when it's spoken than I am at speaking it myself. I tend to take people who are obviously better at this stuff than me (like Tango and MGL and Szym) at their word. So I'm certainly not sure I'm the right guy to try to help you out here, but I'll try.

 

I think the short answer is that correlation coefficients don't mean quite what you think they mean. A few examples -- in these studies, doubles always correlate with runs scored more strongly than triples. I've even seen some where team strikeouts had a slight positive correlation with team runs and team stolen bases had a slight negative correlation. Would it make sense to argue as a consequence of these findings that a double is better than a triple or that striking out is better than stealing a base?

 

As near as I can figure, what Szym was trying to say is that the reason that SLG tends to have a higher r value than OBP at the team level has to do with how the data is distributed. Pearson product-moment correlation coefficients use a working assumption of normal distribution. If you look a little downthread at the Baseball Fever link in my last post, you'll see this (unfortunately clumsily-worded) response from Tango agreeing with Szym's point about team OBP being clustered together (I think) :

 

Also note that there is a lack of understanding of correlation. If all the teams build their teams so that the team OBP is all around the .320 to .340 mark, the regression won't be able to approach 1.00 as well as a teams that are built where HR are clustered.

 

I wish the last phrase didn't have syntax problems that make it impossible to tell exactly what he's saying. So here's my interpretive guesswork: either team OBP violates the normal distribution assumption because there aren't enough outlying datapoints (too clustered in the middle) or (more likely) team SLG violates the normal distribution assumption because there are clusters at various points, not just in the middle. This might be because of park effects -- ballpark dimensions probably tend to have a much more drastic effect on SLG than OBP. This might also feed into strategic roster construction considerations, since there seems to be a tendency to try to build a team to fit the ballpark. Then you end up with too many high-SLG teams built to hit the 3 run HR and too many low SLG teams built to play small ball. If the data isn't normally distributed, you can't trust the correlation coefficients to tell you how much of the variation is explained.

 

In any event, please take this post as a well-intended attempt to collaberatively build a better understanding. I'm sorry if my tone seemed harsh in my previous post. If we have to face down virulent anti-intellectualism to even talk about stuff like this, we might as well think of each other as fellow travelers.

[igor67]

The simplest way to get at the discrepancy is to realize that the correlation coefficient is using straight OBP, whereas the best run estimators have to adjust OBP's value to get the best fit. That's why you can have to seemingly contradictory pieces of information. Also keep in mind that the two statistics are not completely independent. And one thing that could be driving the higher correlation coefficient would be that very high OBP guys tend to also be all of your premier sluggers too. Whereas in the case of say Braun a high SLG doesn't always immediately pull up your OBP.

 

That said and switching gears back to the original thread quote, I think looking at Vlad and even Sammy Sosa can be instructive for Braun career path. Most batters will learn to walk a little bit more as the age, in some cases it can be a very large difference if they are a very talented hitter. I would guess that at this point Braun would be seeing a lot fewer strikes if he wasn't in front of Prince already, but they will have to start throwing outside the zone more sooner or later.

[kramnoj]

A few examples -- in these studies, doubles always correlate with runs scored more strongly than triples. I've even seen some where team strikeouts had a slight positive correlation with team runs and team stolen bases had a slight negative correlation. Would it make sense to argue as a consequence of these findings that a double is better than a triple or that striking out is better than stealing a base?

 

Both of those could make sense to me. Doubles are often a function of power, whereas triples are usually speed based. Power leads to more runs than speed, so if you have a team full of doubles hitters, I think they will score more than a team full of triples (assuming normal levels). As far as the strikeout vs sb, that makes sense to me as well. Teams do a fairly good job of only allowing people who strike out that have power in the bigs. Basically, the swing that leads to power will lead to strikeouts. If you don't strikeout, you're not trying to hit for power. The selection bias that exists within front offices actually works out. Obviously if you go grab a bunch of minor leaguers who strike out without power, that's going to screw things up, but the studies are based on the rosters that major league teams construct.

 

Here's an example of OBP based OPS or slugging heavy OPS. Braun has an OBP of .320 and a slugging of .575 which equals a OPS + of 130. Johnny Damon is having a very good year, with an OBP of .394 and slugging .493 with an OPS+ of 139. Braun slightly edges out Damon in OPS, but because increases in OBP are more valuable than in slugging, Damon is having a more valuable year so far. This was surprising to me when I saw that. I knew that Braun had 20 HR, but didn't know that Damon was on pace to have a career year.

 

To get back to the bbtf thread. It was from a blog post that talked about how the Blue Jays had a decent OBP and were 5th in the league in OBP but only 12th in the league in Runs. That blogger has heard from the stat community that OBP=Runs, and so he says the stat guys are obviously wrong. What this blogger hasn't understood is that power also leads to runs, and the fact that the team is 13th in slugging has a lot more to do with how few runs they've scored than their OBP. That's obviously an extreme example, but I do think that some in the stat community have banged the OBP drum so heavily that the message has become distorted.

[endaround]

It really seems much simpler than that. The variation of OBP across teams is less than the variation of runs scored. So you can't get the necessary variation from OBP alone.

[homer]

Braun walked twice last night so this whole thread is moot. http://forum.brewerfan.net/images/smilies/smile.gif

[sheethead]

I am fascinated by where this thread has gone. I love numbers and finding creative ways to make them more meaningful. But isn't this thread really all about Braun not working the count, taking walks, and getting hit by enough pitches?

[Brawndo the Thirst Mutilat]

I do think that some in the stat community have banged the OBP drum so heavily that the message has become distorted.

 

You have a point there. The other reason the Blue Jays haven't scored any runs is how terrible they've been with RISP, but the stat geek party line on that is "eh, sorry about the rotten luck." Unfortunately, that's not even a thinkable premise in a mainstream world where SC dipwads keep saying "so and so is clutch, and clutch is everything."

[TuesdaysWithRillo]

So would a team of Johnny Damon's do better than a team of Ryan Braun's from an offensive perspective?

[endaround]

Yes, if you mean this year. Damon's RC/27: 7.39. Braun/s RC/27: 6.57

[jjfanec]

Obviously, we all want to see Braun take a couple more walks. My question is, what does Braun's OBP need to be at for him to be considered elite? For some it may be where he is at now, but for the others what are you looking for? Is .350 good enough or does it need to be above .370? Obviously, this is a hard question and maybe worded wrong, because Braun's OBP right now is driven so much by average, so maybe it is simply what is a reasonable walk per plate appearance? Just wanted to get some of your thoughts.

[kramnoj]

One thing that the As did (still might) is that they wanted their minor leaguers to have one walk for every 10 ABs. You couldn't earn a minor league award from the team if you didn't meet that minimum. So if a player gets 600 ABs you would want a player to have about 60 walks. Braun last year had 29 walks in 451 ABs. When I peruse the league leaders in slugging, they often have 80-100 walks a year, and the best have even more. Most of those players will have OBP of .380-.420. Even notorious hacker Kirby Puckett got on base 36% of the time for his career and he was at .370-.379 in his best years.

 

I guess it depends on how few people you describe as elite. Of the top 5 players last year in the NL in OPS+, Fielder's .395 OBP was the lowest. Braun probably needs to be around .400 OBP to be one of the best hitters in the game, even with a slugging over .600.

[Invader3K]

One thing to keep in mind is that Braun has basically only been in the Majors a little more than one season total. He will probably learn to be more patient and draw walks as he matures as a player. I really can't complain about his OBP the way he is hitting, though.

[TuesdaysWithRillo]

Yes, if you mean this year. Damon's RC/27: 7.39. Braun/s RC/27: 6.57

I guess I shouldn't put as much stock as I do in OPS then considering Braun has 16 more points in his OPS at this point. Can you explain to me the formula behind Runs Created? We can take this offline if you want because I'm sure it's been posted somewhere else.

[GenoSeligPrieb]

" So would a team of Johnny Damon's do better than a team of Ryan Braun's from an offensive perspective?" - Tuesdays

 

"Yes, if you mean this year. Damon's RC/27: 7.39. Braun/s RC/27: 6.57" - endaround

 

If baseball were played in a computer simulation, yes. But since it's not, I think 99% of the players, coaches, executives, writers and fans would take a team of Ryan Brauns over a team of Johnny Damons.

 

This is a classic example of a hole we non-Sabremetric types can drive a Hum-Vee through, in the counter argument. Thanks for inadvertently helping us out! http://forum.brewerfan.net/images/smilies/wink.gif

 

The rest of the year, I'd bet real kaysh money that a team of guys who hit like Ryan does would be better suited to keep up the hitting pace, and win a lot more games, than Damon, who is off to a great start, and less likely to sustain this pace of his.

 

Focusing on Ryan Braun's OBP, while he's excelling at his primary offensive duties, is like looking at ESPN's Erin Andrews when she's on camera, and complaining about her voice when she does a sideline report.

[logan82]

TuesdaysWithRillo wrote:

I guess I shouldn't put as much stock as I do in OPS then considering Braun has 16 more points in his OPS at this point.

OPS is a really good stat, but if 2 players are close I would chose the player with the higher OBP. If Braun is around .350 OBP I would be pretty happy considering how high his SLG is. I would obviously prefer him to be around .400 like other elite hitters, but I am not sure he will get that high.

 

He had a .370 OBP last year, but, like others, I thought we would see a dip in his BA and OBP for several reasons. I was hoping some of the dip would be offset by some improvement in plate discipline, but that has not been the case so far. Of course, he may be reading this board considering he has suddenly gotten a bunch of walks.

 

 

I don't use runs created at all and don't know much about it, but from THT

Runs Created. Invented by Bill James, RC is a very good measure of the number of runs a batter truly contributed to his team's offense. The basic formula for RC is OBP*TB, but it has evolved into over fourteen different versions. We use the most complicated version, which includes the impact of hitting well with runners in scoring position, and is adjusted for ballpark impact. RC/G refers to Runs Created Per Game, which Runs Created divided by the number of outs made by the batter, times 27.

[GenoSeligPrieb]

"OPS is a really good stat, but if 2 players are close I would chose the player with the higher OBP. "

 

Depends on the 2 players, I guess, Logan.

 

But don't you consider the offensive roles the 2 players fall into? Say a leadoff type guy has a .400-ish OBP, with a .400 slugging %, and his teammate's a cleanup hitter has a .300-ish OBP with a .500 SLG %. Sure, they both have .800 OPS's, but if your team has a bunch of Ryan Theriots, and not enough Carlos Lees, wouldn't you strive for a better balance?

 

That's why I look at the makeup of the lineup first, before just going for the higher OBP guy, automatically. Just my 2 centavos...

[Brawndo the Thirst Mutilat]

The rest of the year, I'd bet real kaysh money that a team of guys who hit like Ryan does would be better suited to keep up the hitting pace, and win a lot more games, than Damon, who is off to a great start, and less likely to sustain this pace of his.

That's not what TWR asked, or at least that's not the question Endaround heard and was answering. I think you'd be hard pressed to find many (if any) stat geeks that wouldn't agree that Braun is a better bet to hit this well the rest of the season than Damon is. Most of the projection systems had Braun pegged for exactly this kind of season (around 900 OPS), whereas Damon was supposed to be a 760-ish guy. So Damon is overperforming. If that's part of the question, the obvious answer is Braun.

If the question is whose performance so far this year has been worth more runs, given our current best understanding of how runs are produced, the answer is Damon, by a little bit.

Maybe a more interesting question for Brewers fans, given that both guys play for Milwaukee, both are the same age, and projections systems liked them both about as much, is this one (assuming that what we've seen so far in 2008 is perfectly reflective of their true talent hitting, which it isn't):

Would a team of Ryan Brauns score more runs than a team of Prince Fielders?

[TuesdaysWithRillo]

Would a team of Ryan Brauns score more runs than a team of Prince Fielders?

I'd prolly go with Prince on that one because there are more good RHP than LHP in baseball to match up with Braun.