Mock tournaments/seasons using random number generators are something of a pastime around here. The concept started when Carl Coffee ran a fictitious "snowball fight" tournament featuring WSEM pitchers during the 2012 offseason: a fantastic way to fill the downtime that returned with second and third editions the following two off-seasons. The idea also saw use here and there sporadically to mock up partial seasons or the playoffs. The magnum opus for the random number generator concept, though, was the "Fall Fantasy Friendly" series held after the 2012 season. This was a make-believe 4-team league comprised of WSEM players that used the random number generator to decide the outcome of every single plate appearance of a six-game season. Needless to say, this is something we have thoroughly used and abused.
So, how does it work exactly? For the Mock NWLA Tournament, it's a fairly simple model. Each team is assigned a rating out of 100. When two teams meet, the sum of their ratings is entered as the maximum number in the generator; the minimum number is always set to 1. The visiting team takes the lower end of the range, with the home team taking the higher numbers. For example: the Visiting Team has a rating of 60; the Home Team has a rating of 80. If the generator brings up a number between 1 and 60, the Visiting Team gets the win. A number between 61 and 140 coming up would give the win to the Home Team. The possibility of ties in pool play was ignored to keep things as simple as possible.
Pool play did present two additional scenarios we had to handle, though. 1) The pool play draw. Simple enough, and this was done as simple as can be. The generator was set to a max of 4, and we went through the four tiers individually. The four teams in each tier went one-by-one alphabetically: if "1" was generated, they went to Pool A, "2" put them in Pool B, and so on. 2) With 16 teams and only 4 possible records, we needed a tie-breaker for seeding. So, "margin" works well as a stand-in for run-differential. Margin is simply how far the number generated is away from the point where the visiting team and home team numbers are split. If the split is at 75 and the number 60 comes up, the margin is 15. That's a positive margin for the winning side; negative for the losing side. Margin gets tallied over the 3 pool play games for a total. Seeding is then based off the W-L record, first, with margin as the tie-breaker.
So, that's the rundown. The full randomly generated mock tournament results are presented below: pool play draws, individual pool play game results, overall pool play standings and seeding, then finally the double-elimination playoff bracket. Two questions to ponder as you scroll through:
1) Is there a predictive model that better accounts for the bad hop?
2) Now that we have this, do we even need to go to Ohio anymore?
So, how does it work exactly? For the Mock NWLA Tournament, it's a fairly simple model. Each team is assigned a rating out of 100. When two teams meet, the sum of their ratings is entered as the maximum number in the generator; the minimum number is always set to 1. The visiting team takes the lower end of the range, with the home team taking the higher numbers. For example: the Visiting Team has a rating of 60; the Home Team has a rating of 80. If the generator brings up a number between 1 and 60, the Visiting Team gets the win. A number between 61 and 140 coming up would give the win to the Home Team. The possibility of ties in pool play was ignored to keep things as simple as possible.
Pool play did present two additional scenarios we had to handle, though. 1) The pool play draw. Simple enough, and this was done as simple as can be. The generator was set to a max of 4, and we went through the four tiers individually. The four teams in each tier went one-by-one alphabetically: if "1" was generated, they went to Pool A, "2" put them in Pool B, and so on. 2) With 16 teams and only 4 possible records, we needed a tie-breaker for seeding. So, "margin" works well as a stand-in for run-differential. Margin is simply how far the number generated is away from the point where the visiting team and home team numbers are split. If the split is at 75 and the number 60 comes up, the margin is 15. That's a positive margin for the winning side; negative for the losing side. Margin gets tallied over the 3 pool play games for a total. Seeding is then based off the W-L record, first, with margin as the tie-breaker.
So, that's the rundown. The full randomly generated mock tournament results are presented below: pool play draws, individual pool play game results, overall pool play standings and seeding, then finally the double-elimination playoff bracket. Two questions to ponder as you scroll through:
1) Is there a predictive model that better accounts for the bad hop?
2) Now that we have this, do we even need to go to Ohio anymore?
| POOL A | POOL B | POOL C | POOL D |
   
BCW, HFWB,
HVWBL, WWF |
   
HRL, MNWA,
TBW, WSEM |
   
GAWL, LWA,
OCWA, SWBL |
   
BWBL, GBL,
KWL, PWL |
| POOL A - GAME 1 | POOL B - GAME 1 | POOL C - GAME 1 | POOL D - GAME 1 |
|
(+3) defeats
|
(+34) defeats
|
(+25) defeats
|
(+23) defeats
|
| POOL A - GAME 2 | POOL B - GAME 2 | POOL C - GAME 2 | POOL D - GAME 2 |
|
(+17) defeats
|
(+15) defeats
|
(+25) defeats
|
(+1) defeats
|
| POOL A - GAME 3 | POOL B - GAME 3 | POOL C - GAME 3 | POOL D - GAME 3 |
|
(+12) defeats
|
(+11) defeats
|
(+11) defeats
|
(+3) defeats
|
| POOL A - GAME 4 | POOL B - GAME 4 | POOL C - GAME 4 | POOL D - GAME 4 |
|
(+74) defeats
|
(+8) defeats
|
(+55) defeats
|
(+38) defeats
|
| POOL A - GAME 5 | POOL B - GAME 5 | POOL C - GAME 5 | POOL D - GAME 5 |
|
(+7) defeats
|
(+59) defeats
|
(+9) defeats
|
(+44) defeats
|
| POOL A - GAME 6 | POOL B - GAME 6 | POOL C - GAME 6 | POOL D - GAME 6 |
|
(+42) defeats
|
(+2) defeats
|
(+9) defeats
|
(+32) defeats
|
| POOL PLAY RESULTS W-L, +/- Margin (Seed) |
|||
| POOL A | POOL B | POOL C | POOL D |
|
2-1, +47 (2) 2-1, +8 (6) 1-2, +29 (9) 1-2, -84 (15) |
2-1, -2 (7) 2-1, -36 (8) 1-2, +24 (10) 1-2, +14 (12) |
3-0, +45 (1) 2-1, +23 (5) 1-2, +21 (11) 0-3, -29 (16) |
2-1, +30 (3) 2-1, +24 (4) 1-2, -7 (13) 1-2, -47 (14) |
| DOUBLE-ELIMINATION PLAYOFF BRACKET | |||
