THE BOOK--Playing The Percentages In Baseball

Ozzie Guillen, the other Chicago manager, made an even dumber move tonight when he elected to start Dewayne Wise over Nick Swisher in LF. I did a quick projection for each and had Swisher at a .345 wOBA and Wise at a .312 WOBA.

"Rookie pitcher” doesn’t do Samardzija justice. He’s absurdly raw - has great pitches and has no idea where any of them are going. I’m quite frankly amazed that his BB/9 is only 4.5.

Sorry to burst your bubble, but that posting is pretty idiotic.

He walked a left hand hitting .269 to get a righty/righty match up against a guy with exactly 0 hits this year.
To answer your question Chavez has an OBP of .308 this year, and .325 last versus Martinez who has an OBP of .000 this year and .248 last year.

The last thing which is a small thing, but by loading the bases you create a force everywhere.

Men on second and third, 2 outs, the average batting team scores ONE run 5.4% of the time and is held scoreless 72.4% of the time. Bases loaded, 2 outs, they score one run 9.2% of the time and are held scoreless 67.5% of the time.

So Lou was trading the possibility of a big inning for an increased chance of getting out of it unscathed. I guess that would be the gist of his explanation (minus the numbers, of course).

Sorry to burst your bubble, but that posting is pretty idiotic.

I guess you disagree.

You might be right actually.  I didn’t realize that Endy had a high BA projection.  He is probably projected at near .300 versus a RHP.  On the other hand, Ramon probably has near a .300 OBP even against a RHP.  With the bases loaded, as I said, hitting stats go up a lot.  I am not sure about OBP.

So I think it may be near a toss up, in which case I take back what I said about it being a dumb*ss play.  Considering that having the bases loaded is a lot of pressure for a rookie pitcher without a lot of control in the first place, personally I would not have issued the walk.

However, in no way was the post “idiotic.” And in the future, if you call anyone’s posts “idiotic”, you will be banned from this site.  We don’t do that here. This is not “Deadspin.com”

So Lou was trading the possibility of a big inning for an increased chance of getting out of it unscathed.

Dan, according to your numbers, it is the opposite. Loading the bases decreases the chances of being unscathed (67.5 to 72.4).

In any case, you can’t analyze this situation without considering the batters.

BTW, I have always wondered, when a team has a runner on third and no outs in a tie game in the bottom of the 9th or later, why does the pitching team often load the bases?  This was being advocated, or at least the possibility mentioned, by Darling and Hernandez during the Mets broadcast.

You have to play the infield in anyway.  Having the force at home does not really help you in any way.  If you don’t load the bases and play the infield in, the runner does not go on a ground ball unless it gets through anyway.

I suppose there is the possibility of the home to first DP, but that is fairly rare.  That is the only advantage though.  You have to weight that against the “walk loses the game” and the extra pressure on the pitcher.  You also have to consider the batters before you automatically walk the bases loaded.

Looking at the players, with an eye to larger sample sizes…

Samardzija is pretty even in his career l/r splits in the minors, except for bb, only 2.5 per 9 vs rhb, 4.6 vs lhb

Endy’s hitting in the 260s this year, is career .268 vs rhp

Ramon Martinez had spent all of this year in the minors, but has a long mlb career, with a career .315 obp vs rhp

The difference between Endy’s ba and Martinez’s obp is just about the same as the difference in generic REs, about .05

So walking the bases loaded did give the Mets about a 5% better chance of tying the game

Brian that sounds reasonable.  So maybe my criticism was not so idiotic after all.

I also thought it was “curious” that an old-school manager like Lou would elect to put that kind of pressure on a young pitcher. In my book that would certainly be a tie-breaker (not to have him pitch with the bases loaded and the game on the line).

"BTW, I have always wondered, when a team has a runner on third and no outs in a tie game in the bottom of the 9th or later, why does the pitching team often load the bases?”

I presume the fielding team has an easier chance of nailing a force out at home rather than the catcher applying the tag on a close play.

"He’s a young man pitching in a pennant race against some pretty good hitters,” Piniella said of the right-hander. “This is a learning curve for him. The more experience we can get him, the better he’ll be as the season goes on.” (Quote from 08/29/2008, Piniella talking about Samardzija)

So perhaps Lou believes that putting that type of pressure on a young pitcher is what you should do.

Lou said earlier in the week that the rest of the final week was about finding a couple of roles to fill and one of them is the 3rd reliever’s role.  He’s given Howry some high leverage situations again after a couple months without them and is putting Samardzija in them as well.  One of them will get that role.  I don’t think it’s fair to analyze decisions that may be based on other factors that do not include winning today’s game.  However, Lou has shown a tendency to walk the bases loaded when there is an empty base so what he did last night is, at least it seems to me anyway, typical of how he manages.

And in the future, if you call anyone’s posts “idiotic”, you will be banned from this site.  We don’t do that here. This is not “Deadspin.com”

MGL - This is your’s and Tango’s website and you can make any rules that you want.  And I understand that you are just trying to keep all the discussions civil, which is something that I am all for.  But I do think there is a qualitative difference between calling someone an idiot and saying what someone has written is idiotic.  You are certainly correct that there were less inflammatory ways for Dave Rogers to express his disagreement with what you said.  But there were also less inflammatory ways of expressing your opinion than entitling your post “Dumb*ss move by Pinella”.  You could have just as easily written “What was Pinella Thinking?”.  Civillity should extend to all human beings, not just those that we know or who happen to post at this site.  And the tone of the site is set by those who initiate the posts.  Just something to think about.

I suggest a title more like “Questionable move by Piniella”

I think that would be more likely to foster intelligent discussion, which I think is what you want, than calling it a dumba** move, which is more likely to result in the “idiotic post” response.

Those that dish must learn how to take.

Those that dish must learn how to take.

I’m pretty sure MGL was being ironic.  It’s hard to pick up sarcasm in posts.

Dollars to donuts Lou forgot how many outs there were and thought he was setting up a double play.

I think it’s pretty unlikely that Lou forgot how many outs there were in the inning. The guy’s pretty old, but he still has that intense focus during the game.

I think it’s more likely Lou had the scouting report on Martinez, and maybe his CURRENT YEAR stats (I doubt Lou carries around opposing pitcher/batter projections), added it with Lou’s greater familiarity with what kind of batter Endy is and decided it would be better for Shark to face Martinez. Or as someone said above, he wanted to give Shark (I don’t want to spell his real name) a high-leverage situation in preparation for the postseason.

To studes point in #14, I point out a non-baseball blog post from Roger Ebert:

http://blogs.suntimes.com/ebert/2008/09/this_is_the_dawning_of_the_age.html

I don’t believe Lou for a second that he purposely made a move he does not consider “correct” just to give the kid some experience in a pressure situation.  Managers, when they issue the IBB to load the bases, seem to forget that a walk brings in a run and that it is a difficult thing to pitch to a batter with the bags loaded, psychologically and physically (when you also don’t want to walk him), which is why the hitting stats go up.

How often, with runners on 2 and 3 in a tie or one run game, do you hear an announcer (ex-player and/or manager) say, “Well, he has choice between batter A and batter B?” As if they forgot that the choice is between batter A with first base open and batter B with the bags loaded, which is quite different since you are essentially comparing one guy’s BA with the other guy’s OBP and BA adjusted for the bags loaded!

That being said, the Cubs have nothing to play for at this point in time (unlike, say ANA, who is playing for HFA in the ALCS, and BOS and TB who are trying to win the division to avoid playing ANA in the first round and for “bragging rights” I guess), so I give Lou a pass, although if I were a MIL fan, I would have been pretty pissed at that point in time.  It was one of those moments, if you were watching the game, that you just “knew” that he was going to walk Martinez (I don’t mean that literally of course - I would not have bet that he would walk him unless you gave me like 10-1 odds).

I presume the fielding team has an easier chance of nailing a force out at home rather than the catcher applying the tag on a close play.

The point I was trying to make was that if you DON’T walk the bags loaded and play the infield in (which you have to do even with bags loaded and no outs), there IS no play at home on a ground ball because the runner generally does not try for home with the infield in (contact play). 

In fact, with no outs, the runner on third will rarely run unless the ball goes through the infield, for the same reason you don’t put on a squeeze with 0 outs. (With one out, the contact play is sometimes put on, especially with runners on second and third, since even if the guy gets thrown out at home, you still have a runner in scoring position with 2 outs.)

So that argument does not hold any water.

I agree that there is a difference between calling someone an idiot and their post “idiotic.” I also agree that I can be caustic and rude at times, although I have tried to “clean up my act” a little lately.

That being said, I was cordial with that poster and what I said still stands.  Anyone that is excessively rude on this blog will be warned and then banned.  Especially if it is a troll, which this person is/was.  If they are a long-time poster who has contributed a lot of useful content, they will get the benefit of the doubt.

Not to mention the fact that the “owners” of the blog deserve some extra deference.  Not that our views should not be challenged.  Of course they should.  And then some.

Since there is no registration, there is no explicit banning of users. 

We can delete posts (which is extremely rare on this site thankfully), but I agree that if someone starts with “dumb-ss”, it’s hard to think that “idiotic” is much of a step down in the discussion.  Would probably have changed the tone if MGL said “Headscratching move by Lou”.  Then, I don’t think the “idiotic” statement would have been posted.

Stating something as “dumb-ss” and not backing it up with power invited that particular comment.  Indeed, that poster could have removed that line in his comment, and MGL could have made it “Headscratching” instead of “dumbass” and we’d have a perfectly fine thread going here.

There is a big difference between calling a move by a manager a crude name and someone calling someone else’s thoughtful post “idiotic.” I give the troll some credit though as he did support his statement with some data, albeit shaky data (quoting this year and last year’s stats as opposed to some kind of current projection, which is the only thing that counts in order to analyze the decision).  But I don’t think we should dwell on this issue.  I think it was handled appropriately.

In retrospect, the biggest problem with calling the move dumb*ss, was that I reacted a little too strongly and quickly. The move was probably not as bad as I initially thought, although I now think it was clearly wrong.  I do think that what Lou said has some merit (purposely putting pressure on a young pitcher in a meaningless game for the Cubs), however, I still don’t believe that was his reasoning at the time he made the decision. I think he is lying and trying to cover his a**.

I don’t know that Lou is like this in general, but lots of managers seem to think that they have to do “something” in critical situations to show everyone how smart they are.

I work in the database environment, where most companies are always putting out some fire.  The DBA (the guy doing the database work) is always doing something, and as a result are very valuable.

When I went to my Oracle class years ago, the instructor (a DBA from Oracle) said that the DBA that you should hire is the one that never has to put out a fire.  He’s got the operation streamlined well enough that there isn’t always some catastrophe lurking. The problem is that if you look like you are doing nothing, you’re out of a job.

So, it is almost a necessity that the manager make moves. Lots of them.  And as long as the opposing manager is doing equally suboptimal moves, they all end up cancelling out.  “Listen, I give you $1, and you give me 95 cents”.  Later… “Listen I’ll give you back your $1, and you give me back my 95 cents”.  And on and on. 

Also, Felipe Alou once said that the manager he didn’t want to be opposed against was Bobby Valentine.  Not because Bobby V was so good, but that you didn’t know what Bobby would be doing.

The troll or whatever he is was out of line.  I don’t care about the title for this post, but was the post thoughtful?  I’d say no.  It was, as you even said, reactionary. 

As far as what Lou said about finding players for the couple holes that remain, he didn’t say anything after the fact.  I have no clue what he said last night as I usually don’t pay attention to those things, but I do know that Lou said on Sunday or Monday that he had to figure out who was going to valuable to this team in those areas. 

I have no idea what Lou was thinking, and I already said that’s how Lou seems to manage most similar situations, but I just think you can’t analyze what a manager is doing when he has nothing to play for at this point.  If this happened a week ago, I’d probably be on your side here, and would probably have called him out even sooner than you, but last night’s game meant nothing, win or lose.  You must include that in any observations about whether or not he made the right decision.  The fact is that the right decision didn’t matter.  The Cubs have nothing to play for at this point.  Right or wrong, that decision cost the Cubs nothing. 

Whether or not Lou was thinking about that doesn’t matter.  That decision is only wrong if it had a negative impact on the team’s chances and last night may as well have been a spring training game for the Cubs.  It had no impact whatsoever.

As a Mets’ fan, it was the next inning that just killed me.  Murphy hit a triple.  Man on third, no one out, tie game, bottom of the ninth, Wright, Delgado and Beltran due up next.  And the Mets don’t score.

I’m still upset.

I didn’t have complete projections handy when I quoted the numbers, but I did try to go with some longer term to get larger sample sizes. Future plans for my projection system would have vs right and left for batters and pitchers, which along with park factos, would allow calculating matchups - what si the expected production of Endy Chavez vs Samardjia in Shea Stadium?

Guys like LaRussa like to have their prinouts of matchups, but how hard is it going to be to tell them that projections based on larger samples are better than the actual 10 plate appearances that this batter faced this pitcher.

Also, I didn’t figure in a pinch hit penalty for Martinez, assuming that he was a pinch hitter.

I think that Martinez pinch hit earlier, and I don’t know much much a pinch hit penalty a pinch hitter has, if any, in his second PA.

I have been advocating for years that a manager should have “index cards” of all batter/pitcher matchups, but, as Brian says, based on projections. That is completely different, of course, than the matchup cards that managers have now, which are completely worthless (actually less than worthless).

Tonight, Lou mad a similar but troubling, I think, decision.  In the bottom of the 9th, 2 outs and a runner on second, he walked Delgado to pitch to Beltran.  I am assuming that they are similar hitters in that situation.

Of course the correct play is to bring in a lefty pitcher to pitch to Delgado, but Lou is clearly not playing to win (which is another story, and I don’t think that is fair to the Brewers).  However, assuming you don’t, I think that filling up first base puts a lot of pressure on the pitcher, who was a young rookie again (Hart).

Plus, Beltran would need to be a substantially worse batter, since now 2 walks or an infield hit and a walk (etc.) wins the game.

Sure enough, the pitcher got behind on Beltran and threw a bad pitch middle in.  I am sure that if he had gotten behind Delgado with a base open, he would at least have tried not to necessarily throw a strike.

In The Book or in an article he wrote, Andy showed that pitching around a batter with a base open does not seem to have any overall effect.  I have never “believed” that conclusion.  I am not sure that Andy looked at all of the parameters he should have.  In any case, if I am a pitcher (which I was at a fairly high level), I would much rather pitch to a batter with a base open than without a base open, even if that batter is quite a bit better than the next one.

A lot of mistakes made by the Cubs in that game.  The pitch by Howry to Cancel was terrible. He should have thrown a slider (not that you can say that a pitcher “should” throw a certain pitch 100% of the time - that would be bad from a game theory perspective).  The attempted tag at home by Hill was awful.  All he had to do was get up, block the plate and put the tag on.  Instead, he elected to remain on his knees and the runner (Church) slid around him.

There’s a reason Koyie Hill was the starting catcher in AAA this year. By “a” I of course mean “many”.

Thought of this blog when I was watching the Brewers game today.  The game was tied in the 8th inning.  CC Sabathia had pitched 8 innings on 3 days rest and was due to leadoff the bottom of the 8th.  The Brewers decided to let him bat, which seems like one of the most obviously wrong decisions I’ve seen a manager make.  I’m curious what MGL’s sim would say about the Probability of Winning after (a) letting a batter bat then putting in a reliever and (b) letting CC bat and then pitch the 9th.

My guess is that you’re better off with a reliever in the 9th than you are with Sabathia, so that it’s a bad move before you even consider the loss associated with letting the pitcher bat.

Of course, the announcers praised the move, Braun hit his homer and the Brewers did in fact go on to win.  But that can’t have been a good move, no?

Good question.  I was watching the Cubs feed and Brenley, who often has some good insight, actually questioned the move.  In fact, I think he said that he would not have let Sabathia bat.

If it was a run of the mill pitcher (having a good game), it is for sure a pretty bad thing to let him bat.  That fact that it is Sabathia versus a RHP makes it not that bad a play I would think, as he is a good hitter for a pitcher.

Sabathia is an elite pitcher in the NL.  As good as most relievers.  What he is after 115 pitches on 3 days rest, I don’t know.  He sure looked like he was still as sharp as ever.

In this case, I think that Sveum was thinking that if he took Sabathia out and one of his relievers blew the game, he would be lambasted and he would feel awful.  I also think that Sveum already had in his mind that he was going to use Sabathia in that game as long as he was pitching well (which he was) and until his arm fell off.

So while it probably was not the right thing to let him hit, I would say that it may not have been that bad, and it may have even been correct.

Honestly, I am not sure.  I would have to run it through a sim to see the difference between him batting and whoever they had on the bench (including the pinch hitter penalty).  That is the key.  I think you could justify calling it a tossup between Sabathia and a reliever on the mound.

The laughable thing in the game was the Corey Hart AB with the bases loaded and 1 out (7th inning?).  One of the worst AB’s you will ever see. Even though he is a decent hitter overall, he was one of the worst choices in that situation.  He was facing a pitcher with a great frisbee-type slider, which is a devastating pitch against RHB (and not that effective versus RHB). Brenly mentioned before he stepped up to the plate that Hart has been swinging at sliders out of zone all series long.  Plus, Hart hits into a lot of DP’s despite his good speed.

No one even mentioned the possibility of pinch hitting a lefty batter for Hart in this situation.  I don’t know if they even had one on the bench. If managers and coaches are supposed to know things about player strengths and weaknesses beyond the numbers, isn’t this a perfect time for Sveum, like Brenley, to recognize that this is not such a great matchup for the Brewers.  All they needed in that situation was a non-DP ball in play.

In case you didn’t watch the game, Hart swung and missed (by a mile) at 3 straight sliders nowhere near the strike zone.  He looked like he was swinging with his eyes closed.  If you want to get a chuckle, check out the video on MLB.com.  It was a truly pathetic AB.  Interestingly, Hart has horrible career numbers with the bags loaded I think.  We chalk those anomalies up to the vagueries of small samples generally, but maybe there is something about Hart that he chokes in that situation.  He sure looked like a world-class choker in that AB.

I suspect that Sabathia doesn’t have much of an edge over the first bullpen guy, Torres, considering he had pitched 8 innings on short rest.  CC is a horse, but even he has his limits, just look at last year’s playoffs where he was out of gas.

He’s a good hitter for a pitcher, but not the guy you want leading off an inning.  I can understand a manager letting him be a hero, as he’s carried this Brewer team to the brink of the playoffs.  When he was hitting (and taking some big cuts) I was hoping he’d get a hold of one and win the game that way.

But as cool as that seems, as much as I want to let him win the game, I would have summoned some rationality and pinch hit for him.

Sure, pinch hitting is probably the proper thing to do, but I don’t think that Sveum was going to get criticized for leaving him in, even if they lost the game, and we know what often drives a manager’s decision. If he takes him out, and the bullpen implodes, boy is he going to get a lot of backlash.  I think.

As far as Torres or anyone else being better than CC, as we know, pitching is a funny thing.  Last year in the playoffs, he looked like crap.  He threw almost nothing but fastballs, especially when he got behind on the Boston batters (which was often), and they punished him.  From what I have seen this year, at least lately, he has been mixing up his pitches brilliantly.  Honestly, I think that Cleveland had a bad manager, pitching coach, and catcher, but I could be wrong.  Anyway, it sure looked to me like CC was still strong as ever, but I could be wrong about that too.  It really doesn’t look like anyone, including team managers and coaches, know when pitchers have it, lose it, etc. It’s a joke when managers do and don’t take pitchers out.  How often does a pitcher let up ropes, but if they are caught, they are left in the game.  Then as soon as they issue a walk or two bloop hits and a couple of runs score, they are gone.  In the WS game the other day, Vazques was getting shelled.  Ozzie leaves him in there.  Then the next batter gets a check swing accidental double down the line to score 2 or 3 runs and Javy is gone.  Cracked me up.

I mean, why was Sheets even allowed to pitch the day before.  He comes in, pitches like crap for 3 innings and gets relieved.  Couldn’t they see in the bullpen that he wasn’t healthy enough to pitch?  Do you just ask your pitchers, “Hey, can you go tonight?” and if they say, “Yes” then it’s a go?  That’s a good way to run a team!

Seems to me like the catcher would have the best idea whether a pitcher is pitching well or not.  He is the only person who knows if the pitcher is hitting his spots.  Plus he is in the best position to know about the movement on pitches, and he should also be able to judge velocity fairly well.  Because he has to have the trust of the pitchers he works with I can understand why he can come right out and say “get this guy out of here, he doesn;t have anything left tonight”, but it seems like he should be able to pass that information along to a pitching coach or manager in some fashion.

I don’t know what the catcher should know speed and location any more than someone watching the game.  An observer can see where the catcher sets up and what the speed of each pitch is. Plus, on most teams, the manager or pitching coach sends in the signs to the catcher (of course the pitcher can change those).  Plus, catchers are (very flawed) human beings themselves with their own agendas.  Sure, let them give their opinion if they have one, but their opinion is only one small piece of data for the manager and coach to use as far as when to leave pitchers in or take them out. I watch 200-300 games a year and I’ve been doing that for 20 years.  I can categorically say (of course it is just my opinion), that by and large, managers and their pitching coaches have no idea what their pitchers are going to be like going forward in terms of their decisions to take them out or leave them in.  Those decisions are almost always dictated by score and pitch count (and whether the pitcher is going to bat in the NL of course), and other silly things, sometimes ridiculously so (like when they are getting hit hard but letting up no runs, or a few bloop hits cause 5 runs to score).

Ideally, what I would want is for the catcher, manager and pitching coach to relay in their observation, pitch by pitch.  Feed that to a data recorder, so that a data analyst can use that information and see if there’s any predictability in there.

Pitch 102, Santana’s fastball came in straighter than I would have liked, even if the speed is still there.

Pitch 103, I’m only calling for fastballs at this point, since Santana can’t throw a breaking pitch any more.  Fastball had a bit more movement.  Looks good.

Pitch 103, breaking pitch, little bite, lost a bit of speed.  Can’t call it any more.

etc, etc

You feed this data, and then we can know with historical data how reliable their observations are.

I believe in observational data.  I just don’t that the humans can necessarily sift through all the data and past experiene into some wisdom.  And, I think it would be pretty easy to pull off this kind of experiment.  (Catcher could only relay the information by watching video post-game.)

Until then, we can’t distinguish the yappers from those who really know what they are seeing.

It is often not so much that the observations are wrong, but other forces get in the way of reacting to and processing those observations correctly.  As a simply example, and at the risk of repeating myself multiple times, not too many managers are going to take out a pitcher after 5 innings and 80 pitches who is pitching a shutout but is actually pitching like crap (which is entirely possible).  There are dozens of biases, and other forces that get in the way of managers and coaches doing the right thing even when observing correctly.