Forum Archive :
Opening Rolls
For those of you interested I've rolled out all the opening moves top
contenders. Trials 1296 Standard Deviations 0.008. I'm not sure of the
statistical relevance of the data with this S.D. but they,re here for
your perusal. A few points I noticed though were slotting ones own 5
point with 41 is definitely an inferior choice, for 21 its definitely
second place but for 51 seems like a good choice given the general
accepted opinion is that JellyFish doesnt "steer" its games towards
backgames.
Splitting to the opponents 5 point with 43 and 54 produced better
results in the rollouts.
The inner board point making numbers 31 42 53 64 all produced equities
higher than the evaluation.
Top moves marked with *
Dice Move Move L7 EV L6 Roll Dice Move Move L7 EV L6 Roll
21* 13/11 24/23 -.005 +.017 62 24/18 13/11 +.010 +.006
21 13/11 6/5 -.007 -.008 62* 24/16 -.003 +.008
21 24/21 -.018 -.008 62 13/5 -.020 -.015
21 13/10 -.026 -.027 62 24/18 24/22 -.041 -.042
21 24/22 24/23 -.035 -.056
43 13/9 24/21 +.010 -.003
31* 8/5 6/5 +.140 +.155 43 13/9 13/10 +.005 -.004
43* 24/20 13/10 +.001 +.009
41* 13/9 24/23 -.010 -.006 43 24/20 24/21 -.007 +.003
41 13/8 -.022 -.037
41 24/20 24/23 -.026 -.023 53* 8/3 6/3 +.052 +.077
41 13/9 6/5 -.032 -.031 53 13/8 24/21 +.029 +.024
41 24/20 6/5 -.048 -.066 53 13/8 13/10 +.008 -.003
51* 13/8 24/23 +.008 +.004 63* 24/18 13/10 +.002 .000
51 13/8 6/5 -.014 -.004 63* 24/15 +.002 .000
51 24/18 -.025 -031 63 24/18 24/21 -.031 -.047
63 13/4 -.039 -.037
61* 13/7 8/7 +.098 +.107
54* 13/8 24/20 +.025 +.034
32* 24/21 13/11 +.002 +.011 54 13/8 13/9 +.022 +.015
32 13/10 13/11 -.005 +.004 54 24/15 +.002 .000
32 13/10 24/22 -.007 -.008
32 13/8 -.022 -.037 64 24/14 +.014 +.007
32 24/21 24/22 -.027 -.027 64 24/18 13/9 +.011 +.008
32 13/10 6/4 -.049 -.038 64* 8/2 6/2 +.005 +.011
42* 8/4 6/4 +.095 +.122 65* 24/13 +.079 +.066
52 13/8 24/22 +.010 +.001
52* 13/8 13/11 +.008 +.007
52 13/8 6/4 -.030 -.041
52 13/6 -.040 -.054
|
|
Chuck Bower writes:
Midas wrote:
> For those of you interested I've rolled out all the opening moves top
> contenders. Trials 1296 Standard Deviations 0.008. I'm not sure of the
> statistical relevance of the data...
I can shed some light on this. Take the difference between two
plays (of the same roll!) and compare to the following table:
difference confidence level
0.014 90%
0.017 95%
0.026 99%
This is based on the 0.008 standard deviation quoted. What "confidence
level" really means is:
If you could do a VERY LARGE NUMBER of rollouts ("VERY LARGE NUMBER"
means the biggest number you can think of), what are the chances that
the order of the two plays will remain the same?
If the difference between two plays is less than 0.015, then statistically
you can call them equal. (0.015 is not a HARD number. Some might say
0.02 and others 0.01. But at some point you have to draw the line
statistically, and the range 0.01 to 0.02 for Midas's table should satisfy
most statisticians, I suspect.)
The other question is: "How well is JF playing these positions?"
Given that JFv3.0 level-6 is playing its twin, it is fair to say that
the two players are equal! But for any given position, the "defender"
may have an easier position than the "offender", or vice versa, so the
rollout results could have a "systematic error". These are difficult
to estimate. As a wild guess, I'd say 0.01 for these opening roll
positions. All the more reason to consider two plays "equal" if they
are within a certain distance of each other, that distance being around
0.02, give or take a bit, IMHO.
Chuck
bower@bigbang.astro.indiana.edu
c_ray on FIBS
|
|
|
|
Opening Rolls
- At different match scores (Louis Nardy Pillards, July 2002)
- Average advantage of winning opening roll (Chuck Bower, Oct 1998)
- Choosing a strategy (Daniel Murphy, June 2001)
- Early game rule of thumb (Rich Munitz, Feb 2009)
- Factors to consider (Kit Woolsey, July 1994)
- How computers play (Kit Woolsey, Mar 1995)
- Magriel's Chapter 5 (Hayden Alfano+, May 2006)
- Mloner vs Jellyfish (Kit Woolsey, Dec 1995)
- Nactating a whole game (Nack Ballard+, Jan 2011)
- Nactation (Jim Stutz+, June 2010)
- Nactation overview (Nack Ballard, Oct 2009)
- Nactation--Why use it? (leobueno+, Jan 2011)
- Opening 1's: Split or slot? (Douglas Zare, Dec 2003)
- Opening 21: Rollout (Stick, Mar 2006)
- Opening 21: Split or slot? (Dick Adams+, Dec 2003)
- Opening 32: Rollout (Stick, Feb 2006)
- Opening 43: In GOL online match (Raccoon+, Feb 2004)
- Opening 43: Pros and cons (Stick+, Jan 2006)
- Opening 43: Which split is better? (Peter Backgren+, Aug 2000)
- Opening 43: Which split is better? (Michael J. Zehr+, Mar 1996)
- Opening 51: Rollout (Stick, Feb 2006)
- Opening 52: Merits of splitting (Peter Bell, Apr 1995)
- Opening 53: Magriel's recommendation (George Parker+, July 1997)
- Opening 53: Split to 21? (Alex Zamanian, Aug 2000)
- Opening 53: Why make the three point? (Kit Woolsey+, Feb 1996)
- Opening 6's: Slot the bar point? (Chuck Bower+, Feb 2000)
- Opening 6's: Slot the bar point? (David Montgomery, June 1995)
- Opening 62: Could running be best? (Gary Wong, Sept 1997)
- Opening 62: Split, run, or slot? (Chuck Bower, May 1997)
- Opening 63: Middle Eastern split? (Mark+, Apr 2002)
- Opening 63: Slot the four point? (Dennis Cartwright+, Mar 2002)
- Opening 64: Make the two point? (William Hill+, Jan 1998)
- Opening 64: Make the two point? (Darse Billings, Feb 1995)
- Opening 64: Rollout (Peter Grotrian, Jan 2006)
- Opening 64: Split to 20? (Peter Bell, June 1995)
- Opening 64: Three choices (Brian Sheppard, July 1997)
- Opening 65: Becker on lover's leap (Jeffrey Spiegler+, Aug 1991)
- Opening 65: Computer rankings (Chuck Bower, Jan 1997)
- Opening rolls ranked (Arthur+, Apr 2005)
- Rollouts of opening 21 and replies (Alexander Nitschke, Oct 1997)
- Rollouts of openings (Tom Keith+, Jan 2006)
- Rollouts: Expert Backgammon (Tom Fahland, Aug 1994)
- Rollouts: Jellyfish 3.0 (Midas+, Sept 1997)
- Rollouts: Jellyfish 3.0 level 6 (Chuck Bower, Feb 1999)
- Rollouts: Snowie 4.1 (Rene Cerutti, Apr 2004)
- Slotting the four point (Joe Loria+, Oct 1999)
- Snowie's openers and replies (rcerutti, Feb 1999)
- Splitting versus building (Dave Slayton+, Aug 2000)
- Splitting versus slotting (Daniel Murphy, Apr 2001)
- Splitting versus slotting (Daniel Murphy, Sept 1997)
- Trice's rankings (Marty Storer, Feb 1992)
From GammOnLine
Long message
Recommended reading
Recent addition
|
| |
|