Here's a great substitution...

http://onlinegames.strat-o-matic.com/game/346899/1

I sent this to Strat this morning.

Why do Carpenter and Johnson exchange positions?

Mainly because the switch improved your team defense. 3b(x) comes up 50% more times than 1b(x), so HAL corrected your defensive mistake.

You went from:

1b-3e16
3b-4e25

to

1b-4e25
3b-3e20

Actually, I don't agree, Apovlika. The substituted fielding is slightly worse:

This is a simplified analysis just using singles and total error chances (in decimal format) from the fielding X chart:

Original lineup positions:

Carp. 1b, single .20, error .287, total .487, times 2 die roll chances = total chances = .974 die roll chances (out of 216)
Johnson 3b , single .30, error .301, total .601, times 3 die roll chances = total chances = 1.803 die roll chances

Total die roll chances for hit/error for set lineup, without considering compound results - 2.777 out of 216

Revised lineup positions:

Carp. 3b, single .20, error .241, total .441, times 3 die roll chances = total chances = 1.323 die roll chances (out of 216)
Johnson 1b , single .30, error .449, total .749, times 2 die roll chances = total chances = 1.498 die roll chances

Total die roll chances for hit/error for set lineup, without considering compound results - 2.821 out of 216

Inferiority of substituted positons 2.821 - 2.777 = .044

Inferiority as a percent of total chances from original lineup - .044/2.777 = 1.6%

These probabilities are from DeanTSC's (? MeanDean's) charts for X chart probabilities, especially re the error chances.

By the way, the compound results and 2 base error results make the inferiority worse - it goes up to about 2.7% when those effects are considered.

Questions:
-are my calculations correct? (I think they are)
-whether HAL should be switching positions for a relatively minor net fielding effect, especially when the positions are not their primary positions.
-are there other ways of looking at this? Apovlika - if you calculated it to this detail, would you have done it differently? (I'm curious).

Geoff

Geoff, I think you're right on the money. In addition to that, the 1bman loses one range rating every time a runner is held on 1B, so that 4 becomes a horrible 5 in those cases!

Strat's response...

"Thank you for bringing this defensive switch to our attention. In our latest defensive upgrade we had only prevented corner outfielders from switching positions as this was the most common bug that had occurred. Carpenter and Johnson had switched here as Johnson at 1B and Carpenter at 3B is actually a (very slightly) superior defense than Carpenter at 1B and Johnson at 3B since there are more x-chances for 3B than 1B. We will work on preventing the game from switching other players other than corner outfielders in the future."

Well, they provided a response. That's always a good thing. I think there is a message that this is something that they're going to do more work on.

To put a positive spin on this, I know Bernie used to say that it was really hard to make changes in the old TSN environment. At least now this is something that they are working on, and that changes can be made, and are being made.

Now, if Dean's charts could wend their way into their web consciousness......

Whether it was the right or wrong move doesn't bother me Geoff.

The fact that it was done WITHOUT my knowledge bothers me. I'd like to be able to be in charge of these moves.

Hell, I won the game that it occurred in. I just don't like moves being made when they are not suppose to.

Whether it was the right or wrong move doesn't bother me Geoff.

The fact that it was done WITHOUT my knowledge bothers me. I'd like to be able to be in charge of these moves.

Hell, I won the game that it occurred in. I just don't like moves being made when they are not suppose to.

I agree.

Interesting. I understand Visick's concern. ........ I think the move did improve the defense though... Of course I could be wrong. I want to read what gbrookes wrote again. Anyone know the definitive answer for this? I am almost certain it makes the defense slightly better, which was SOM's response. . But back to Visick's concern, he should be able to control that. However, most of us are vets maybe a newer GM wants Hal to improve his defense??... It is interesting to think about, thanks for posting.

From my calculations with the original lineups

at 3b you have a 30% chance of a single, and a 30% chance of an error on 3 of 216 chances = 1.80
at 1b you have a 20% chance of a single, and a 29% chance of an error on 2 of 216 chances = 0.98

with hals changes

at 3b you have a 20% chance of a single, and a 24% chance of an error on 3 of 216 chances = 1.32
at 1b you have a 30% chance of a single, and a 45% chance of an error on 2 of 216 chances = 1.50

either way you have a +30% chance of a single at 1b with runner held--so I see no difference in odds there

so I rate it 2.78 1st way, 2.82 second way (this is just compared to each other, not odds)--the original lineup is very slightly better to me (almost insignificant really), but with a runner at 1st, with possible DPs it may swing the other way.