2010-04-07 Rockies (Cook) at Brewers (Davis), 12:10 PM CDT [Brewers win, 5-4]

[ELCABALLO45]

Geez rluz,

 

I know you don't agree with what happened during the game but not everything is played to the book. Someone is just trying to explain his stance to you that not everyone thought Gomez bunting was a bad idea (me included) but you really don't need to sound so conceded towards the play or towards anyone who disagrees, just saying.

 

Anyways, GREAT job by the bullpen today. Nice to see Hoffman get another save. Always great to win a series, always great to win a series against a good team and always great to win a series when you lose your first game.

 

2-1. Bring on St. Louis

[rluzinski]

I already said this, and anytime in any baseball game when you are losing by 1 the value of 1 run is great.

 

I think you still aren't appreciating how the inning of the game influences the relative value of a run. The run differential of the game is only one factor to consider.

 

I hate the added probability stats a lot, they are pointless and meaningless to me on the scale of the team W-L. For individual players it is ok. Seriously, obviously if you are winning 4-0 in the first you have a good chance to win but not amazing, but winning 4-0 in the 9th you have an amazing chance to win, wow, Im so glad someone tracks this for me. Infact, I did not have the intuition to realize that going from a 1 run to a 2 run lead in the 6th inning increases your chances of winning but not by a huge margin, but if you went from a 1 run lead to a 7 run lead in the 6th inning your chances increase a lot...amazing! The fact the people try to claim that they can calcualte these effects to more precisely than single digits is laughable.

 

I tried to answer your question, “Are runs worth less in the early innings?” Why did you ask it if the answer is so obvious?

 

And I was speaking in generalities when I said “win probability”. A manager has to make an educated guess at the win probabilities of the various strategies at his disposal so he can try to select those strategies that will result in a win more often than not. If you think that is "pointless and meaningless" than arguing about baseball at all is as well.

And they can calculate the probabilities to the nearest digit, provided every other factor is exactly the same. They obviously aren’t though, so we must do the best that we can to adjust for them. To throw your hands up in the air and say, ”But we can’t possibly adjust for every factor, so the numbers are worthless!” simply isn’t true. The numbers are the best numbers we have. If I have imperfect evidence supporting one strategy and no evidence supporting the other, I know what I’m doing.

 

For the rest of it, I think much of what you said is wrong but I don't want to write anymore in the gameday thread. If you are interested in continuing this discussion, please start a thread in the major league or sabermetric forum.

[topper09er]

5 runs/game run environment... sacrifice!

 

It's easy to sit there and point out what factors might make our estimates less accurate but you are offering nothing more accurate to use instead. No estimate is perfect; we just make our best educated guess.

Last year in the ML the average runs per game was 4.6 and even lower at 4.4 in the NL. The brewers averaged 4.8 runs/game, so if those are reliable numbers then the brewers record should have been 162-0. Well, thats silly because no statistical model is perfect so Im sure the record was not 162-0, it was something close. Actually, they were 80-82 which is pretty far away, that is weird, that must mean that using the average runs to predict the future outcome is not a good idea. In fact, for the 6 NL central teams last year, the average runs were 4.3/game with a standard deviation of 3.0, and a COV of .70. That is huge and basically means that you should assume that average values are not good representatives of a random event from the sample.

 

So it looks like instead of just trying to score 5 runs every single game it is better to make decisions based on game to game situations to maximize your wins.

 

[RobDeer 45]

Boom roasted

[rluzinski]

5 runs/game run environment... sacrifice!

 

It's easy to sit there and point out what factors might make our estimates less accurate but you are offering nothing more accurate to use instead. No estimate is perfect; we just make our best educated guess.

Last year in the ML the average runs per game was 4.6 and even lower at 4.4 in the NL. The brewers averaged 4.8 runs/game, so if those are reliable numbers then the brewers record should have been 162-0. Well, thats silly because no statistical model is perfect so Im sure the record was not 162-0, it was something close. Actually, they were 80-82 which is pretty far away, that is weird, that must mean that using the average runs to predict the future outcome is not a good idea. In fact, for the 6 NL central teams last year, the average runs were 4.3/game with a standard deviation of 3.0, and a COV of .70. That is huge and basically means that you should assume that average values are not good representatives of a random event from the sample.

 

So it looks like instead of just trying to score 5 runs every single game it is better to make decisions based on game to game situations to maximize your wins.

Obviously, I meant that you should use the average runs/game scored in the majors over the last 10 years for all in game decisions. You wouldn't want to base in game decisions on the expected runs for a particular match up and update that as the game goes on.

 

I can see this is a waste of time.

[topper09er]

So what data do you have to show the added probability of Carlos Gomez bunting Rickie Weeks to 3rd base with 1 out with Braun/Fielder coming up facing Aaron Cook with the score 3-4 in the 5th inning at Miller park in a day game factoring in the Rockies lineup/bench/bullpen and the Brewers lineup/bench/bullpen vs allowing him to hit away there? Because earlier you stated that it doesnt matter what happened there you know it was the wrong play because based on games from 1999-2002 that play would cause the brewers to score less runs in the game.

 

rluzinski wrote:

---

When discussing optimal game strategy, it's pretty irrelevant how it turned out in one instance. We have thousands of games of data that suggests it is generally a bad move to sacrifice bunt in that kind of situation. If Gomez was on his own and thought he could get a hit, so be it. I can't imagine he caught the corners napping, though.

 

[sdcheezhd]

Thanks Topper; the Braun stats are helpful although you still need more info to conclude how much a sacrifice increases the chances of scoring 1 run if Braun comes up next as compared to the overall 13% average. I wonder what the sample size is for innings with a guy on 2d and 0 out and no sac ahead of Braun vs. innings with a guy on 2d and 0 out and a sac ahead of Braun. And then you'd have to decide how to treat situations where the batter ahead of him is trying to advance the runner by hitting to the right side. Maybe that would be slicing and dicing too much.

 

On the SF in the 9th of the 1st game, I guess I didn't focus on that b/c in the 9th down 3 that's a fail, not a productive run-scoring out (or hit). And even if you count it it makes him 1-3 on the year, not 1-2.

[phxMILWfan]

I love calling him Diego. It allows me to be a fan of his and still hate Jim Edmonds. If only he would wear a disguise. Mr. Potato head style glasses and mustache.

[ilovebeer]

I had a great time at the game yesterday, I took my mother and 6 year old nephew, we parked on at slims on bluemound and took there shuttle. The only bad thing about yesterday was waiting for the bus in the cold and windy rain,, the rain was coming in sideways and it actually hurt. We had to wait for about 10 minutes. I never got warm the rest of the day.

[SeriesFinale]

It was a treat finally having MLB Network and being able to watch this game on the FSN Wisconsin feed in actual HD as an out-of-market fan.

[Invader3K]

I had a great time at the game yesterday, I took my mother and 6 year old nephew, we parked on at slims on bluemound and took there shuttle. The only bad thing about yesterday was waiting for the bus in the cold and windy rain,, the rain was coming in sideways and it actually hurt. We had to wait for about 10 minutes. I never got warm the rest of the day.

Yes, walking out of the stadium in freezing rain was not a nice feeling. Thankfully I wore layers and a fairly thick jacket, but my face felt like it was being sandblasted. Plus hobbling around on a sprained ankle didn't make it any better. It was a great game, though, which made up for any suffering outdoors.

[zzzmanwitz]

I hate off days...