 |
View Old Forum Thread
Old Forum Index » Baseball Beta Testing » Beta News » Sacrifice Bunts
tysonlowery
I still need to add the fielders to the bunts (who is making the plays on these bunts). That will come next.
Sacrifice Bunts
July 23, 2002 at 09:47PM View BBCode
Your pitchers will now attempt sacrifice bunts in obvious bunting situations (man on first or second, with nobody on third, and less than 2 outs). The thought being that a pitcher isn't skilled enough to pull off the squeeze.I still need to add the fielders to the bunts (who is making the plays on these bunts). That will come next.
celamantia
July 24, 2002 at 12:36AM View BBCode
Instead of just limiting this to pitchers, you may want to include position players who's bunt ratings are higher than the average of their contact and power ratings.
tysonlowery
July 24, 2002 at 01:20AM View BBCode
That might be a good way to do it. Although I'm not sure if some managers would want to bunt with those guys. What do other people think?
hcboomer
July 24, 2002 at 01:38AM View BBCode
Not sure how many players that would apply to, but unless it was Rey Ordonez I don't think I'd want any position guy always bunting in "obvious" bunting situations.
celamantia
Apply the following formula:
BB-(BP*2)-BC+((NP*2+NC)/3)+((RS*2+BS)/3)
Where:
BB=Batter's Bunt Ability
BP=Batter's Power
BC=Batter's Contact
NP=Next Batter's Power
NC=Next Batter's Contact
RS=Runner's speed
BS=Batter's Speed
The result of this formula is the percentage chance of attempting a sacrifice bunt. Ideally, it would also take into account the ranges of the opposing pitcher, catcher, and corner infielders; say, average those ratings together, invert them (so 80 becomes 20) and weight the result of the formula above accordingly: (BF*2+RF)/3, where BF is the bunt formula above and RF is the range formula (100-((CR+1BR+3BR+PR)/4)). Thus, the final formula would be:
((BB-(BP*2)-BC+((NP*2+NC)/3)+((RS*2+BS)/3))*2+(100-((CR+1BR+3BR+PR)/4))) /3
where:
BB=Batter's Bunt Ability
BP=Batter's Power
BC=Batter's Contact
NP=Next Batter's Power
NC=Next Batter's Contact
RS=Runner's speed
BS=Batter's Speed
CR=Opposing Catcher's Range
1BR=Opposing 1B's Range
3BR=Opposing 3B's Range
PR=Opposing Pitcher's Range
The result of this formula, again, is the chance of attempting a sacrifice bunt with a position player with a man on first or second, with nobody on third, and less than 2 outs.
Whaddaya think?
July 24, 2002 at 01:49AM View BBCode
How's this:Apply the following formula:
BB-(BP*2)-BC+((NP*2+NC)/3)+((RS*2+BS)/3)
Where:
BB=Batter's Bunt Ability
BP=Batter's Power
BC=Batter's Contact
NP=Next Batter's Power
NC=Next Batter's Contact
RS=Runner's speed
BS=Batter's Speed
The result of this formula is the percentage chance of attempting a sacrifice bunt. Ideally, it would also take into account the ranges of the opposing pitcher, catcher, and corner infielders; say, average those ratings together, invert them (so 80 becomes 20) and weight the result of the formula above accordingly: (BF*2+RF)/3, where BF is the bunt formula above and RF is the range formula (100-((CR+1BR+3BR+PR)/4)). Thus, the final formula would be:
((BB-(BP*2)-BC+((NP*2+NC)/3)+((RS*2+BS)/3))*2+(100-((CR+1BR+3BR+PR)/4))) /3
where:
BB=Batter's Bunt Ability
BP=Batter's Power
BC=Batter's Contact
NP=Next Batter's Power
NC=Next Batter's Contact
RS=Runner's speed
BS=Batter's Speed
CR=Opposing Catcher's Range
1BR=Opposing 1B's Range
3BR=Opposing 3B's Range
PR=Opposing Pitcher's Range
The result of this formula, again, is the chance of attempting a sacrifice bunt with a position player with a man on first or second, with nobody on third, and less than 2 outs.
Whaddaya think?
celamantia
Now, if the batter's bunt rating is 80, and the next batter's power is 75, and all of the other numbers stay the same, the chance jumps to 47.78%. However, if that same batter also has an 85% power rating, the chance drops to a measly 1.1%. If, on the other hand, the average range of the four opposing fielders is ony 30 instead of 50, the chance goes up to 7.78%.
July 24, 2002 at 02:00AM View BBCode
As a baseline, if every number in that formula was average (50), the bunt would be called for 16.67% of the time, or once out of every six times the situation comes up.Now, if the batter's bunt rating is 80, and the next batter's power is 75, and all of the other numbers stay the same, the chance jumps to 47.78%. However, if that same batter also has an 85% power rating, the chance drops to a measly 1.1%. If, on the other hand, the average range of the four opposing fielders is ony 30 instead of 50, the chance goes up to 7.78%.
celamantia
July 24, 2002 at 02:11AM View BBCode
In fact, using this formula, you don't need to distinguish between pitchers or position players at all. Since most pitchers have D to F offensive ratings, they'll naturally have a high result on average. For example, a pitcher with 50 speed, 20 power, 20 contact, and 50 bunt, with the next batter having 70 power and 60 contact, and the runner having a 50 speed, and an average defense (50 range across the board) will go for the bunt 87.78% of the time.
tysonlowery
What if we worked the score in somehow? If the score is within one, and its late in the game, the chances of bunting should go up.
July 24, 2002 at 02:46AM View BBCode
I like this idea. Whereas with pitchers you always attempt, with batters its based on the skills involved.What if we worked the score in somehow? If the score is within one, and its late in the game, the chances of bunting should go up.
celamantia
0.5+((Inning*Inning)/(30*((ABS(HomeScore-VisitorScore)*(ABS(HomeScore-VisitorScore)/2)+4)/2)))
This takes into account score differences:
Difference 3 runs, 4th inning: Multiplier is 0.625
Difference 1 run, 6th inning: Multiplier is 1.03
Difference 6 runs, 9th inning: Multiplier is .75
Difference 2 runs, 9th inning: Multiplier is 1.4
Difference 1 run, 9th inning: Multiplier is 1.7
Tie score, 11th inning: Multiplier is 2.52
Which makes the complete formula:
(((BB-(BP*2)-BC+((NP*2+NC)/3)+((RS*2+BS)/3))*2+(100-((CR+1BR+3BR+PR)/4))) /3)*(0.5+((Inning*Inning)/(30*((ABS(HomeScore-VisitorScore)*(ABS(HomeScore-VisitorScore)/2)+4)/2))))
where:
BB=Batter's Bunt Ability
BP=Batter's Power
BC=Batter's Contact
NP=Next Batter's Power
NC=Next Batter's Contact
RS=Runner's speed
BS=Batter's Speed
CR=Opposing Catcher's Range
1BR=Opposing 1B's Range
3BR=Opposing 3B's Range
PR=Opposing Pitcher's Range
July 24, 2002 at 03:47AM View BBCode
Okay, try this formula as a multiplier to the previous formula:0.5+((Inning*Inning)/(30*((ABS(HomeScore-VisitorScore)*(ABS(HomeScore-VisitorScore)/2)+4)/2)))
This takes into account score differences:
Difference 3 runs, 4th inning: Multiplier is 0.625
Difference 1 run, 6th inning: Multiplier is 1.03
Difference 6 runs, 9th inning: Multiplier is .75
Difference 2 runs, 9th inning: Multiplier is 1.4
Difference 1 run, 9th inning: Multiplier is 1.7
Tie score, 11th inning: Multiplier is 2.52
Which makes the complete formula:
(((BB-(BP*2)-BC+((NP*2+NC)/3)+((RS*2+BS)/3))*2+(100-((CR+1BR+3BR+PR)/4))) /3)*(0.5+((Inning*Inning)/(30*((ABS(HomeScore-VisitorScore)*(ABS(HomeScore-VisitorScore)/2)+4)/2))))
where:
BB=Batter's Bunt Ability
BP=Batter's Power
BC=Batter's Contact
NP=Next Batter's Power
NC=Next Batter's Contact
RS=Runner's speed
BS=Batter's Speed
CR=Opposing Catcher's Range
1BR=Opposing 1B's Range
3BR=Opposing 3B's Range
PR=Opposing Pitcher's Range
tysok
If the score is tight and late with the number 8 hitter up, he won't bunt to bring the pitcher up. However, if you have (for some reason) a 300+ hitter on the bench he possibly would.
The number 8 hitter usually won't bunt, the #3, 4 and 5 is rare. 6 and 7 do occasionally, but the majority is the 1, 2 and 9 hitter... if 9 is your pitcher. :)
Also you may have a solid 300+ hitting #8 to stack a little more power to the bottom of the lineup. Or he's there for strategy reasons (that are nonexistent so far)... as in the #7 has great power but only hits low 200, a strong able #8 would act as a barrier... perhaps getting better pitches for #7.
I think the idea of having the score (and other factors) involved in some situational logic is a thing long since overdue. The logic behind which pitcher comes into relieve has always been a problem, as well as who (and occasionally when to) pinch hit.
Also, I don't have time to analyze Celemantia's formula, but the next batters ability to get on base (contact for singles) should figure in. Why give up an out if the guy has 500+ OBP? Then you'd need to account for the next hitter as well... That could become a nightmare. :)
About pitchers, they don't always bunt the guy over to 3rd, chances of the double play are reduced as long as no ones on 1B, see what the pitcher can do and if he Ks you've got the next batter with a man in scoring position anyhow... bunting to 3B isn't always automatic, the speed of the man on 2B makes some kind of difference in the decision.
July 24, 2002 at 03:56AM View BBCode
Following Tysons post about bringing the score into the situation, you also need to look at the other hitters, or who's on the bench.If the score is tight and late with the number 8 hitter up, he won't bunt to bring the pitcher up. However, if you have (for some reason) a 300+ hitter on the bench he possibly would.
The number 8 hitter usually won't bunt, the #3, 4 and 5 is rare. 6 and 7 do occasionally, but the majority is the 1, 2 and 9 hitter... if 9 is your pitcher. :)
Also you may have a solid 300+ hitting #8 to stack a little more power to the bottom of the lineup. Or he's there for strategy reasons (that are nonexistent so far)... as in the #7 has great power but only hits low 200, a strong able #8 would act as a barrier... perhaps getting better pitches for #7.
I think the idea of having the score (and other factors) involved in some situational logic is a thing long since overdue. The logic behind which pitcher comes into relieve has always been a problem, as well as who (and occasionally when to) pinch hit.
Also, I don't have time to analyze Celemantia's formula, but the next batters ability to get on base (contact for singles) should figure in. Why give up an out if the guy has 500+ OBP? Then you'd need to account for the next hitter as well... That could become a nightmare. :)
About pitchers, they don't always bunt the guy over to 3rd, chances of the double play are reduced as long as no ones on 1B, see what the pitcher can do and if he Ks you've got the next batter with a man in scoring position anyhow... bunting to 3B isn't always automatic, the speed of the man on 2B makes some kind of difference in the decision.
celamantia
I don't take batting average or OBP into account because those can be misleading early in the season and for players that don't play every day.
July 24, 2002 at 04:16AM View BBCode
The next batter's contact and power are taken into account (with weight given to power; should that be reversed?), but not the bench, since there's no guarantee that ABE will actually bring in the pinch hitter at that point.I don't take batting average or OBP into account because those can be misleading early in the season and for players that don't play every day.
hcboomer
July 24, 2002 at 01:51PM View BBCode
This whole issue is one of those things that will be helped if and when some form of managerial tendencies are added to the game. That way everyone can decide for themselves how frequently they want to bunt and tweak it as necessary depending on how it seems to be working.
celamantia
There are two parts to this that managers should be able to take into account:
- Who tries for bunts
- How often bunts are tried for
The first affects the first equation. Say this is a rating from 0 to 10. Multiply this rating by .2, then multiply the result with the batter's bunt ability in the equation. This will make batters more or less likely to bunt, as it increases the weight of bunt ability over hitting ability.
For how often the manager goes to the bunt, this affects the entire equation. Again, this should be a rating from 0 to 10. Again, multiply by .2, then multiply the results of the entire equation against this value. A 0 here will prevent sacrifice bunts from EVER happening!
The only hard part about this is figuring out how to word it on the screen!
July 24, 2002 at 02:21PM View BBCode
This would be extremely easy to add to this equation. In fact, you can add the feature now and this can be the first one in!There are two parts to this that managers should be able to take into account:
- Who tries for bunts
- How often bunts are tried for
The first affects the first equation. Say this is a rating from 0 to 10. Multiply this rating by .2, then multiply the result with the batter's bunt ability in the equation. This will make batters more or less likely to bunt, as it increases the weight of bunt ability over hitting ability.
For how often the manager goes to the bunt, this affects the entire equation. Again, this should be a rating from 0 to 10. Again, multiply by .2, then multiply the results of the entire equation against this value. A 0 here will prevent sacrifice bunts from EVER happening!
The only hard part about this is figuring out how to word it on the screen!
tysok
Man on 1B with 1 out, top of 9th down 2 runs. Speed of A.
Batter has contact of A, power of D, speed of A, AVG is 305, OBP is 400.
The next batter has contact C, power A, speed D, AVG is 220, OBP is 280.
The next batter has contact A, power B, speed B, AVG is 280, OBP is 350.
Then the pitcher.
Everyone has a bunt ability of B just for easy reference. :) Also I'll guess an A=70 pnts, B=60 pnts, C=50 pnts, D=40 pnts.
All the fielders are C.
According to the formula (correct me if I'm wrong, I think I got the numbers plugged in right) you'll sacrifice the runner to 2nd 68.4% of the time. Now you'd have 2 outs with a guy on 2nd, and a bad hitter coming up. The visiting team has to play for the win, but chances are they won't even get the tie.
I would assume what would be done is let the batter swing away, try to get 2 men on, then bunt with the next man. to get 2 runners in scoring position for batter #3 to get in.
One thing I noticed when working with this is as the batters running speed goes down, the chance of bunting also goes down... All things being equal except speed of A and speed of C you would more likely bunt with a C speed, to try an stay away from the double play, although I've also noticed in this game some of the fastest runners are the ones hitting into more double plays (another problem in my opinion).
I think the formula needs to SUBTRACT the batters speed, instead of adding it. This ends up making the likelyhood of a sacrifice higher for a B rated runner than an A rated runner (the numbers I used gave me a 24.8 for an A runner and a 28 for a B runner).
With the same numbers in the scenario above, and the subtraction of the batters speed, that drops the sacrifice number to 24.89%. Still to high I think.
If you were to reverse the contact and power, the batter would never sacrifice the runner to 2nd (-3). Given those same circumstances, but take out the current batter... making batter #2 the current batter, the batter would then sacrifice 15% of the time. Which makes sense overall.
If you put batter #2 on base, and made batter #3 the current batter, he would never sacrifice (-68).
I'm going to play a bit more with the formula as I get some more time. I'd like to plug in some different circumstances and see how they work out. With those 2 changes it looks like a fairly solid formula, although I'd like to review it some more and make sure no variables are missing or something is weighted too much etc....
How you came up with this extremely long and involved formula is beyond me, made my head hurt for about 10 minutes while I tried to decipher it. :)
July 24, 2002 at 03:51PM View BBCode
Cela: Aye. Take this scenario.Man on 1B with 1 out, top of 9th down 2 runs. Speed of A.
Batter has contact of A, power of D, speed of A, AVG is 305, OBP is 400.
The next batter has contact C, power A, speed D, AVG is 220, OBP is 280.
The next batter has contact A, power B, speed B, AVG is 280, OBP is 350.
Then the pitcher.
Everyone has a bunt ability of B just for easy reference. :) Also I'll guess an A=70 pnts, B=60 pnts, C=50 pnts, D=40 pnts.
All the fielders are C.
According to the formula (correct me if I'm wrong, I think I got the numbers plugged in right) you'll sacrifice the runner to 2nd 68.4% of the time. Now you'd have 2 outs with a guy on 2nd, and a bad hitter coming up. The visiting team has to play for the win, but chances are they won't even get the tie.
I would assume what would be done is let the batter swing away, try to get 2 men on, then bunt with the next man. to get 2 runners in scoring position for batter #3 to get in.
One thing I noticed when working with this is as the batters running speed goes down, the chance of bunting also goes down... All things being equal except speed of A and speed of C you would more likely bunt with a C speed, to try an stay away from the double play, although I've also noticed in this game some of the fastest runners are the ones hitting into more double plays (another problem in my opinion).
I think the formula needs to SUBTRACT the batters speed, instead of adding it. This ends up making the likelyhood of a sacrifice higher for a B rated runner than an A rated runner (the numbers I used gave me a 24.8 for an A runner and a 28 for a B runner).
With the same numbers in the scenario above, and the subtraction of the batters speed, that drops the sacrifice number to 24.89%. Still to high I think.
If you were to reverse the contact and power, the batter would never sacrifice the runner to 2nd (-3). Given those same circumstances, but take out the current batter... making batter #2 the current batter, the batter would then sacrifice 15% of the time. Which makes sense overall.
If you put batter #2 on base, and made batter #3 the current batter, he would never sacrifice (-68).
I'm going to play a bit more with the formula as I get some more time. I'd like to plug in some different circumstances and see how they work out. With those 2 changes it looks like a fairly solid formula, although I'd like to review it some more and make sure no variables are missing or something is weighted too much etc....
How you came up with this extremely long and involved formula is beyond me, made my head hurt for about 10 minutes while I tried to decipher it. :)
tysok
I would agree that setting your own managerial tendencies would be very cool, and a necessary part of the game in the future. However at this point I would argue against it.
Right now all managers are identical, we can look at a game and see that our best hitters are sacrificing, and try and find a fix. Whereas if there were options on the tendency we may not know how much of that problem is based on the formulas in the game, or if the problem is in the formulas taking care of the tendencies... As more tendencies get added in (do you go for extra bases or play the short game?) it could get bogged down to the point where finding the problem having to do with your best hitter bunting could be in 4 to 10 different formulas.
I would rather see the changes take place slowly over a period of time so the problems can be worked out, and then the tendencies added after everything has been balanced correctly.
Weighting the bunt ability by .2 or .4 may be slight, but it could possibly result in a solid A contact hitter sacrificing leaving the heavy work for a a D contact A power hitter...
July 24, 2002 at 04:00PM View BBCode
New posts while I was typing. :)I would agree that setting your own managerial tendencies would be very cool, and a necessary part of the game in the future. However at this point I would argue against it.
Right now all managers are identical, we can look at a game and see that our best hitters are sacrificing, and try and find a fix. Whereas if there were options on the tendency we may not know how much of that problem is based on the formulas in the game, or if the problem is in the formulas taking care of the tendencies... As more tendencies get added in (do you go for extra bases or play the short game?) it could get bogged down to the point where finding the problem having to do with your best hitter bunting could be in 4 to 10 different formulas.
I would rather see the changes take place slowly over a period of time so the problems can be worked out, and then the tendencies added after everything has been balanced correctly.
Weighting the bunt ability by .2 or .4 may be slight, but it could possibly result in a solid A contact hitter sacrificing leaving the heavy work for a a D contact A power hitter...
celamantia
Unfortunately, I left my test spreadsheet at home. I'm going home for lunch, though , so I'll plug in your changes, your test case, and the numbers ffrom the other bunt thread.
--Chris
July 24, 2002 at 04:27PM View BBCode
You definately have better baseball knowledge than I do, Tysok, so I'll defer to you on those changes. I think I need to change a constant somewhere to balance it out, though.Unfortunately, I left my test spreadsheet at home. I'm going home for lunch, though , so I'll plug in your changes, your test case, and the numbers ffrom the other bunt thread.
--Chris
celamantia
Well, I couldn't go home for lunch, but I've recreated my spreadsheet, and unless I've got my own formula wrong, the problem is worse than you suspect. For this situation, I'm showing a 144% chance of a bunt attempt!
Changing that to a subtract on my spreadsheet drops this to a 14% chance.
On my spreadsheet, reversing the weight between contact and power in this situation shows -88% chance; the bunt will never happen.
Now, let's plug in the numbers from the other thread to this formula with your changes.
In his first example, which was the pitcher (Johnny Benjamin), the formula says Benjamin is 306% likely to bunt. Good call.
He was lifted, though, for a pinch hitter who shouldn't bunt, Hod Morgan. Plugging in Hod Morgan's numbers, we have a -403% chance of a bunt, again a good call.
Although these two calls are good, I think the changes you suggested need to be balanced out. In the old situation, woith everything equal with a 1 run difference in the 6th inning (my "base scenario"), the old formula had a 16% chance of a bunt; in the new formula, it's -51%.
Tysok, you're far better at this sort of thing than me. Can you think up two or three other scenarios and how you think the batter should go, so I can use that as a comparison point?
--Chris
July 24, 2002 at 06:13PM View BBCode
Originally posted by tysok
Cela: Aye. Take this scenario.
Man on 1B with 1 out, top of 9th down 2 runs. Speed of A.
Batter has contact of A, power of D, speed of A, AVG is 305, OBP is 400.
The next batter has contact C, power A, speed D, AVG is 220, OBP is 280.
The next batter has contact A, power B, speed B, AVG is 280, OBP is 350.
Then the pitcher.
Everyone has a bunt ability of B just for easy reference. :) Also I'll guess an A=70 pnts, B=60 pnts, C=50 pnts, D=40 pnts.
All the fielders are C.
According to the formula (correct me if I'm wrong, I think I got the numbers plugged in right) you'll sacrifice the runner to 2nd 68.4% of the time. Now you'd have 2 outs with a guy on 2nd, and a bad hitter coming up. The visiting team has to play for the win, but chances are they won't even get the tie.
Well, I couldn't go home for lunch, but I've recreated my spreadsheet, and unless I've got my own formula wrong, the problem is worse than you suspect. For this situation, I'm showing a 144% chance of a bunt attempt!
I would assume what would be done is let the batter swing away, try to get 2 men on, then bunt with the next man. to get 2 runners in scoring position for batter #3 to get in.
...
I think the formula needs to SUBTRACT the batters speed, instead of adding it. This ends up making the likelyhood of a sacrifice higher for a B rated runner than an A rated runner (the numbers I used gave me a 24.8 for an A runner and a 28 for a B runner).
With the same numbers in the scenario above, and the subtraction of the batters speed, that drops the sacrifice number to 24.89%. Still to high I think.
Changing that to a subtract on my spreadsheet drops this to a 14% chance.
If you were to reverse the contact and power, the batter would never sacrifice the runner to 2nd (-3). Given those same circumstances, but take out the current batter... making batter #2 the current batter, the batter would then sacrifice 15% of the time. Which makes sense overall.
On my spreadsheet, reversing the weight between contact and power in this situation shows -88% chance; the bunt will never happen.
Now, let's plug in the numbers from the other thread to this formula with your changes.
In his first example, which was the pitcher (Johnny Benjamin), the formula says Benjamin is 306% likely to bunt. Good call.
He was lifted, though, for a pinch hitter who shouldn't bunt, Hod Morgan. Plugging in Hod Morgan's numbers, we have a -403% chance of a bunt, again a good call.
Although these two calls are good, I think the changes you suggested need to be balanced out. In the old situation, woith everything equal with a 1 run difference in the 6th inning (my "base scenario"), the old formula had a 16% chance of a bunt; in the new formula, it's -51%.
Tysok, you're far better at this sort of thing than me. Can you think up two or three other scenarios and how you think the batter should go, so I can use that as a comparison point?
--Chris
celamantia
In his first example, which was the pitcher (Johnny Benjamin), the formula says Benjamin is 211% likely to bunt. Good call.
He was lifted, though, for a pinch hitter who shouldn't bunt, Hod Morgan. Plugging in Hod Morgan's numbers, we have a -100% chance of a bunt, again a good call.
Using Tysok's earlier scenario, we get a bunt chance of 72%. Uh, oh... not good.
It looks like I need to take number of outs into account, and possibly look more than one batter ahead. The next after the guy on-deck is the pitcher, but in the top of the 9th he'll probably get pulled for a pinch-hitter if there are runners on base and 2 outs.
Back to the drawing board...
July 24, 2002 at 06:51PM View BBCode
Okay, I've reworked the formula; now, arms are also taken into account.In his first example, which was the pitcher (Johnny Benjamin), the formula says Benjamin is 211% likely to bunt. Good call.
He was lifted, though, for a pinch hitter who shouldn't bunt, Hod Morgan. Plugging in Hod Morgan's numbers, we have a -100% chance of a bunt, again a good call.
Using Tysok's earlier scenario, we get a bunt chance of 72%. Uh, oh... not good.
It looks like I need to take number of outs into account, and possibly look more than one batter ahead. The next after the guy on-deck is the pitcher, but in the top of the 9th he'll probably get pulled for a pinch-hitter if there are runners on base and 2 outs.
Back to the drawing board...
tysonlowery
We can start with assuming managers have a 50 for sacrifice bunts. Then after we're fairly happy with the results, I can add a screen in so that managers can adjust the number specifically for them.
July 24, 2002 at 08:16PM View BBCode
Sounds like you two are on the way to figuring it out! If you can get a reasonable forumula, it should be fairly easy to plug in to my code.We can start with assuming managers have a 50 for sacrifice bunts. Then after we're fairly happy with the results, I can add a screen in so that managers can adjust the number specifically for them.
tysok
For this purpose, I'm not sure some things are needed in the equation. I haven't had enough time to thoroughly test my ideas, or even the formula Cele put up, but I thought I would take the time to put them in here so I could see if this makes sense to the rest of you.
Deciding whether or not to try and sacrifice the runner over I don't think you'd take fielders range or arm into account. Also I wouldn't look at a guys power, I'M trying to move a man into scoring position so the next hitter can hit him in, I could just leave him there if I cared about a Homer. I'm also not sure how much care I would give to the running speeds of the players, The runner should have an easy time crawling there, and the batter is trying to get out anyhow.
I do think contact is important, taking into account the batter as well as the next one or two batters (you don't want to sacrifice just to bring up a worse hitter than who's up now). Bunting ability is obviously a consideration (you'd prefer to have the ball actually bunted). Also the score and the number of outs are important (no need to sacrifice with two out, and if you're 2 runs down there's no point in sacrificing the runner to 2nd or 3rd if it's in the 9th or later).
((((BB/(1+#OUTS))-((BC+(BC-NC)*(3-#OUTS))/(1+#OUTS))+((NC*(1.1-#OUTTS))/(1+#OUTS)))/3)*1.4)-#OUTS
This is a premature formula I've scratched out by modeling off Cele's first. It doesn't take into account where the runners are, and I'm not even sure it needs to be that complex. :)
This uses 1.4 multiplier, as if it were the ninth inning and 6 to 5 according to Cele's other formula. I figured the best way to take the runners position into account would be in the multiplier formula.
This seems to take the bunt ability, contact of the batter and next batter, and the number of outs into account.
With 2 outs only an F or D hitter will decide to try and bunt, that should be fixed by not even having the formula run if there are 2 outs. :)
Also in the test numbers I have... If there are 0 outs, only a C or lower hitter will sacrifice (and only to bring up 2+better hitter). With 1 out only a C or lower will sac. (and only to bring up 2+better hitter).
2+better hitter means this:
A hitter = 70 pnts.
B hitter = 60 pnts.
C hitter = 50 pnts.
D =40
F =30
No one will sac just to bring up the next best. For instance an F won't sac just to bring up a D, but he will to bring up a C. The same as you go up, D will sac for a B and C will sac for an A.
I know that the multiplier is going to make a big difference in the equation, for instance a C hitter with a B up next with 1 out in the above equation equals -9.6 (he won't bunt). But with the multiplier being 2.4 it becomes -8.6 (still won't bunt but it's higher negative). If the multiplier becomes 15.4 in this he will bunt, =4.4
The way it's working out I think if it comes up positive he should bunt, negative or 0 and he won't.
Ideally you would make this guy bunt if you were down by 1 in the 9th. Haven't even looked at the multiplier formula yet, I'm sure both formulas need tweaking to take the different variables into account... mainly want to know what others think are important variables?
July 25, 2002 at 11:52PM View BBCode
It seems like the chance of a successful sacrifice is already in place. So what we're doing is working out a way to decide if you do or don't try.For this purpose, I'm not sure some things are needed in the equation. I haven't had enough time to thoroughly test my ideas, or even the formula Cele put up, but I thought I would take the time to put them in here so I could see if this makes sense to the rest of you.
Deciding whether or not to try and sacrifice the runner over I don't think you'd take fielders range or arm into account. Also I wouldn't look at a guys power, I'M trying to move a man into scoring position so the next hitter can hit him in, I could just leave him there if I cared about a Homer. I'm also not sure how much care I would give to the running speeds of the players, The runner should have an easy time crawling there, and the batter is trying to get out anyhow.
I do think contact is important, taking into account the batter as well as the next one or two batters (you don't want to sacrifice just to bring up a worse hitter than who's up now). Bunting ability is obviously a consideration (you'd prefer to have the ball actually bunted). Also the score and the number of outs are important (no need to sacrifice with two out, and if you're 2 runs down there's no point in sacrificing the runner to 2nd or 3rd if it's in the 9th or later).
((((BB/(1+#OUTS))-((BC+(BC-NC)*(3-#OUTS))/(1+#OUTS))+((NC*(1.1-#OUTTS))/(1+#OUTS)))/3)*1.4)-#OUTS
This is a premature formula I've scratched out by modeling off Cele's first. It doesn't take into account where the runners are, and I'm not even sure it needs to be that complex. :)
This uses 1.4 multiplier, as if it were the ninth inning and 6 to 5 according to Cele's other formula. I figured the best way to take the runners position into account would be in the multiplier formula.
This seems to take the bunt ability, contact of the batter and next batter, and the number of outs into account.
With 2 outs only an F or D hitter will decide to try and bunt, that should be fixed by not even having the formula run if there are 2 outs. :)
Also in the test numbers I have... If there are 0 outs, only a C or lower hitter will sacrifice (and only to bring up 2+better hitter). With 1 out only a C or lower will sac. (and only to bring up 2+better hitter).
2+better hitter means this:
A hitter = 70 pnts.
B hitter = 60 pnts.
C hitter = 50 pnts.
D =40
F =30
No one will sac just to bring up the next best. For instance an F won't sac just to bring up a D, but he will to bring up a C. The same as you go up, D will sac for a B and C will sac for an A.
I know that the multiplier is going to make a big difference in the equation, for instance a C hitter with a B up next with 1 out in the above equation equals -9.6 (he won't bunt). But with the multiplier being 2.4 it becomes -8.6 (still won't bunt but it's higher negative). If the multiplier becomes 15.4 in this he will bunt, =4.4
The way it's working out I think if it comes up positive he should bunt, negative or 0 and he won't.
Ideally you would make this guy bunt if you were down by 1 in the 9th. Haven't even looked at the multiplier formula yet, I'm sure both formulas need tweaking to take the different variables into account... mainly want to know what others think are important variables?
Pages: 1