In-game_Strategy
In-game_Strategy
Wednesday, May 28, 2014
According to Table 126 of The Book, in the bottom of the 9th inning, the only time that you should put the winning run on base is if all the following is true:
(a) you have at least one runner on base, with 1B is open
(b) you have two outs
(c) the batter at the plate generates about 50% more runs than the batter on deck
So, never with the bases empty. That chapter was written by Andy. When I did my "When to walk Bonds" charts ten years ago, there was a situation where you WOULD walk Bonds with the bases empty. That would be with two outs, and in the bottom of the 8th, 9th, or top of the 9th.
Bonds however generated runs at double the batter on deck, and it's possible that Andy didn't consider such an extreme example in his chart. Mark Reynolds' career wRC+ is about double that of Yovani Gallardo. So, it is possible that Reynolds is to Gallardo what Bonds is to the average hitter.
Except my charts presume that the rest of the batters that inning are also average hitter. In addition, my charts required two outs.
So, in order to understand if the move was justifiable, we have to do a bit more work to figure it out. Because this is the bottom of the 9th, then we don't need to run a simulation. You can figure it out using prob&stats. Hopefully, an aspiring saberist among the Straight Arrow readers will impress us.
?
Thursday, May 22, 2014
Interesting article and good data from Jonah and Dan, along with some insights from the Cardinals pitchers and coach.?
Other than inertial reasoning.
At the very least, why not a designated pinch hitter (meaning after the one at bat, the hitter is out of the game, while the pitcher remains in the game)?
Why MUST the substitution rules be that once a sub comes in, then you must come out? So, simply change the substitution rules, and never utter the word "DH". This will make all NL fans happy.?
Friday, May 16, 2014
MGL has an excellent list for consideration.
Friday, May 09, 2014
Poz channels his inner Bill James and comes up with a system to quantify how he feels when he sees each intentional walk.
One of the things I learned from Bill James is that alot of what we do is the organization of ideas. Take for example what we did in The Book to identify relievers who had nothing left in the tank. It wasn't important that we get something precise that captures every instance. We just need to figure out a way to "bucket" things so that we can disproportionately represent what we need to represent. My system was straightforward enough, and that was to take the batters faced today and add 80% of the number of batters faced the prior day and add 60% two days ago and 40% three days ago and 20% four days ago. I called it Tank points or something. I just needed some reasonable way to track who was likely to have the least left in the Tank.
Your typical Regressionistas would have come up with something more elaborate, go through some mathematical gyrations that no one could possibly understand, and you'll find that the simple system gets us 90% of the way there anyway.
Another example is Shutdowns and Meltdowns. Saves, Holds, and Blown Holds are crappy stats. They are so crappy that they contain no useful information that you can't get better elsewhere. Since these stats are dancing around the idea of WPA, why not just use WPA and to a lesser extent RE24 (or even WPA/LI)? WPA of course are kind of non-intuitive, or at least, not easy to talk about, especially with decimals flying around. But, to simply bucket the changes into an outcome like Shutdown or Meltdown? Yes, that does a good job of giving a high-level view of a reliever's impact.
So it goes with the IBB, to try to identify which IBB are more troublesome. We could of course do something more elaborate, tracking win expectancy with that batter and the on-deck batter (which I have done in the past with Barry Bonds) and what Andy has in The Book. But really, other than me and Andy, who even looks at those charts more than a couple of times? Poz however synthesizes the main parts of the objectionable parts of the IBB into a simple system. Will he capture ALL the bad IBB and ensure he doesn't misclassify any of the good ones? No, of course not.? But, it will disproportionately capture the bad IBB.
Poz gives out the opening salvo. Someone out there can try to improve upon it, but I think it's a pretty good start already.
Monday, May 05, 2014
I certainly won't go into the prediction game. The chance that Ortiz will be the last full-time DH is probably under 10%. You have first of all the younger Billy Butler. So, even if it were to end between the two, Ortiz is probably at 50/50 at best. Then of course, MLB is an ongoing entity, and having an all-hit no-field hitter is an inevitability. Could such a player get, say, at least 5000 PA at DH?
Anyway, setting that aside, this is just my recurring plea to unify the rules on DH. I understand the following:
- Some people love inertial reasoning. So, I won't be able to talk to them.
- Some people love that there are different rules between the leagues. I can understand, appreciate, and support this idea. But to have it limited to just ONE rule? And for that rule to be the DH? I mean, why not go crazy and have something like NFL and CFL type rules? One league has 110 yards and the other is 100 yards. Can't you do the equivalent in MLB if you really are interested in differing rules? One league has 4 downs and the other has 3. Again, can't you change something more specific than just the DH rule if you are so interested in different rules? Let each league establish distance between each base, or the count, or the outs per inning, or the innings. Oh, that's too drastic? So, you like the differing rules as long as it's not CFL/NFL different. To me, the argument is just rationalization. And I won't be able to talk to them either.
So, to the extent that the AL and NL have decided to consolidate just about every other rule in the book, we may as well close off this one. The question is how best to do it. And my favorite suggestion was the "free market" approach. Rather than impose how to handle the DH onto all 30 teams, have each team decide for their home park on the rule. Whether that is before each season, or before each series, or before each game, I'm all ears as to whatever works best.
We ALREADY have this rule. We have the home-park rule for each conference, when we have inter-?conference games. And in spring training, you can have managers confer and agree on the DH. So, we already have something like it. We just need to remove the arbitrariness of the conference as the deciding point, and put the onus on each team to make the decision.
Saturday, April 26, 2014
?Bryan is tracking every single bunt in the NL.
Thursday, April 24, 2014
The commenters at Fangraphs are suggesting he hit as a RHH!?
Friday, April 18, 2014
And Ryan Howard has a pretty big career platoon split to boot. Pretty bold move.?
Thursday, April 17, 2014
In one of the comments here, the reader notes that in Mark Buehrle games, the game time is 2:37. That's 157 minutes. But, that's based on Buehrle pitching one-third of the time, his bullpen pitching one-sixth, and his opponents pitching one-half.
In non-Buehrle games, let's say the average game time is 180 minutes. We follow simple math:
157 = Buehrle * 1/3 + League * 2/3
Since League = 180, that sets Buehrle to 111 minutes. What does that represent? If all pitchers pitched at Buehrle's pace, a baseball game would last 111 minutes. That's 1:51.
Now, I have not verified any of my numbers. Therefore, as aspiring saberist can figure out the right numbers (weighted by year-IP), and give us Buehrle's actual pace.
Then, I'd love for someone to find the slowest pitcher, and tell us how slow his games would be, if all pitchers pitched as slowly as he did.
?
Wednesday, April 16, 2014
Good stuff from Ben. He notes that between 2010 and 2014, the number of seconds between pitches is up by about 0.85 seconds. With nearly 300 pitches per game, that's over 4 minutes. Game time is up by 12 1/2 minutes.? So, it's not just the time between pitches.
Fangraphs has the Pace figures for all players, including in this handy leaderboard.
Sunday, April 06, 2014
?Interesting that the discussion got into the nuances of the ball-strike count.
Friday, April 04, 2014
A straight arrow reader reminded me of this solution we had. I don't remember what our ultimate opinion was, so, I'll throw it back out there again. I also sent it to Poz and Bill, and maybe they'll talk about it as well.
One of the Straight Arrow readers had proposed that: in any 4-pitch walk (or hit batter), the batting team has the option to send the on-deck hitter to first base, and keep the batter at the plate to start at 0-0. (If you must, you can limit it to 1 or 2 outs, or if you really must: 2 outs.)
Basically, trade spots with the on-deck hitter.
So, for those teams that like the idea of "setting up the DP", they can still do it.
For teams to walk the #8 hitter to face the pitcher: that will stop tomorrow.
For teams to walk Cabrera (even unintentionally, but without a strike) to face whoever is on deck: well, now Cabrera is guaranteed to get either at least one strike, or have at least a runner on base.
Let's just say this is payback for all the times he was handcuffed.
?
Friday, March 21, 2014
?Jeff takes a look.
Friday, February 28, 2014
Jason on Fangraphs rolls up his sleeves, and does terrific work in looking into each Rangers bunt to see how it matches to The Book. The aspiring saberist out there should do it for every team, and see how each manager ranks.
***
As an aside, in reading the comments to that article, I have to keep reminding people: we do not, EVER, care about scoring "exactly" 1 run. It is ALWAYS about "at least" 1 run.
This is much better understood if you focus on win expectancy (WE) and not run expectancy (RE), then we wouldn't ever be having this at least / exactly discussion. The Book provides those WE tables, so, I'd like to see those cited instead of RE. RE is a useful proxy, but when readers get confused with the at least / exactly 1 run, then we should simply talk about WE.
Thursday, February 20, 2014
?These guys not only claim to have a model that highly predicts when a manager will pull a pitcher, they also claim their method for pulling pitchers is better. PDF is here.
Monday, February 17, 2014
?His hitting coach with the Yankees said this:
[quote] “He overcame so much while he was here,’’ Long said. “As a young kid there were holes everywhere. There were holes in his swing, in his makeup, in his body composition. This kid grew and grew and grew.
“All the other stuff … he’d take plays off in the field, he’d give away at-bats in RBI situations. He made a lot of personal decisions to get over the hump in those areas. People don’t know how hard he worked, how many times he was the one asking me to do extra work in the cage.’’[/quote]
Fortunately, we have a pretty good stat that tells us how a hitter hits in various base-out situations. It's called RE24 (run expectancy by the 24 base-out state), which you can find at both Fangraphs, and Baseball Reference. Since BR has the data I need in one nice little chart, I'll reference that one. What we do is compare his actual performance in the 24 base-out states, to his overall performance. That is, compare his RE24 to his "BtRuns". And from 2005-2009, his situational performance was 56 runs lower than his overall performance. So, there's definitely a strong indication that he did not tailor his performance to the base-out situation in his youth. Since then, his situational performance was 17 runs higher than his overall performance.
Whether all this was a conscious decision, or simply luck, you guys can look into it more.
***
As for his lack of hustle tainting him: as I've said in the past, at the minimum, you should run out as hard as you would in practice. That's because, at the very least, you would treat the game as if it was practice. It's NOT a question as to whether running at 60% or 90% will add bases or runs or wins to THIS appearance. No, that's a strawman. It's simply the exact same idea as to why you run at 90% or higher in practice: to develop good habits so that when it DOES matter, you will be running at 100% in the future.
Basically, if it came to an actual real-game close play, who is more likely to have run at 100%: the guy who always runs at 100%, or the guy who picks and chooses when to run at 100%?
That said, what counts in the end is your performance. You don't knock Gretzky down because he didn't finish his checks. You don't knock MVP candidate Vlad down because he chased pitches. Not unless you think these things are ultimately indicators of a poor future.
So, don't knock Cano's supposed lack of hustle for the same reason. Especially since is the last guy who got a bunt double.
Tuesday, February 11, 2014
?We've talked about this a few times on the old blog (due a search for Catfish, McDowell, a few others). Neil brings it up today.
I don't remember what we had concluded, but going through it quickly: there's some 20 or 30 points of wOBA of advantage you'd gain, and if it's for two plate appearances, that's worth about 0.04 runs each inning. Fielding-wise, you have to figure you've got a Frank Thomas kind of fielding player, which means it'll cost you some 20 to 30 runs per 162 G on the field to knock out your 1B, or an average of under 0.02 runs per inning.
The other aspect is the injury factor of making someone play an unfamiliar position, but presumably they'd have plenty of practice beforehand.
Naturally, if you are going to knock out your 1B from the game, it better be the 9th inning, and the 1B's turn at bat is not due to come up. So, you have to also figure out the chance that the game will go into extra innings without your 1B batting. Someone else can take it from here...
Friday, January 03, 2014
?I like the question and love the approach.
Approaching it from the other angle, the authors of The Book found that relievers "can handle a much heavier workload than current managers are imposing" without diminishing their performance. This suggests that bullpens may not need as much saving as is usually assumed.
Working off this research, I wanted to determine the dynamic impact a longer-than-average start has on a team's winning percentage in subsequent games. If, as is claimed, a starter who pitches longer into games refreshes the bullpen, we would expect to see the performance of a team's bullpen improve, translating into a greater-than-expected winning percentage in the few days after a long start. And if this effect exists, it should be included in our calculations of pitcher value, as it suggests that an "innings-eater" -- those starters who can be counted on to pitch a large number of innings, year after year -- adds value to his team not only on the days he pitches but also on the next few days.
?An interesting article that missed one important item, best captured here:
Looking at the all-time IBB leaders, their average WPA per IBB cluster around that number: .0107 for Hank Aaron, .0100 for Willie McCovey, .0102 for Manny Ramirez, .0106 for Ken Griffey, .0101 for Prince Fielder, .0105 for Albert Pujols, .0111 for Miguel Cabrera. Vladimir Guerrero is the far outlier, at .0120 per walk. Bonds, at his peak, nearly doubled that.
WPA assumes "average" future conditions, and average present conditions. A great hitter, like Aaron, Junior, Pujols, etc, are worth, on average, about +.07 runs per PA. But in IBB situations (runner on base, 1B open), it's probably more like +.10 runs per PA, which is +.01 wins per PA. And, lookie there, the WPA of their IBB was +.01 wins per PA. This is why we call IBB win-neutral: it's not that their win value is 0, but that their win value relative to the other options (pitching to a great batter) is win-neutral.
So, when the author states that Bonds doubled their numbers at his peak, well, I think that's fine. After all, at his peak, 2001-2004, generated +.17 runs per PA in all situations, which means that in IBB situations, he would have otherwise generated +.23 runs per PA, or +.023 wins per PA. That he was issued IBB below this level, on average, likely shows that many, most, of his IBB were good IBB.
But what about that bases loaded one? I talked about it here.
Page 3 of 5 pages
< 1 2 3 4 5 >
Recent comments
Older comments
Page 2 of 150 pages < 1 2 3 4 > Last ›Complete Archive – By Category
Complete Archive – By Date
FORUM TOPICS
Jul 12 15:22 MarcelsApr 16 14:31 Pitch Count Estimators
Mar 12 16:30 Appendix to THE BOOK - THE GORY DETAILS
Jan 29 09:41 NFL Overtime Idea
Jan 22 14:48 Weighting Years for NFL Player Projections
Jan 21 09:18 positional runs in pythagenpat
Oct 20 15:57 DRS: FG vs. BB-Ref
Apr 12 09:43 What if baseball was like survivor? You are eliminated ...
Nov 24 09:57 Win Attribution to offense, pitching, and fielding at the game level (prototype method)
Jul 13 10:20 How to watch great past games without spoilers