Originally Posted by Stephen Flamm
The best solution to the scoring issues is simply to allow all qualifiers to compete in every event. There were only 45 men and 45 women; with intelligent programming and good organization, each athlete could have completed each event. I think they have all earned the right to do so.
This would seem ideal. The problem, like last year, is fairly judging 16 people is much easier than 45 people without doing multiple heats and simpler events. Still, I think the games organizers really wanted to have all 45 till the end. Then again, is it really necessary?
Please realize, CF Games are claiming that a pool of 3000 men that tried out at the sectionals, that led to the 500 regional, the 45 CF games (4 events), the 24 second sets, the 16 third set was a fair way to distill the most fittest man currently on the planet. That's pretty darn good considering you don't have a clue as to what to do until a couple hours ahead of the events.
Now, the next best thing would have been let all 3000 sectional members compete at the games for a more fair estimation of who's more fit. Sadly, that's just not feasible in the current set up.
So is the scoring of the games fair? Should the range be reset to account for the smaller pool? I don't think so as it doesn't punish those that score REALLY low when competing in a large pool of competitors (the better gauge of fitness after all).
However, I ran an experiment on spreadsheet: What if the average event score for a person's current rank AFTER a cut was used to offer points for that position from there on out. Reason being, it made sense that these guys were finishing in the exact same order among the 45 guys. It accounted for how a guy is 12th place at that point was doing so that should be the new score for 12th place after that. Most like the top 16 would not score around the top 45th place guy at the cut had he been allowed to play. If you look at the spread on the sheet you'll notice that the top 24 are not scoring badly on the individual events (maybe one of them scored a 39).
Anyway, how it's done is consider after the FOUR scored events was the first cut, so you divide by 4 for the average:
1. Graham Holmberg (18) - 4.5
2. Rich Froning Jr. (23) - 5.75
3. Matt Chan (24) - 6.0
4. Chris Spealler (37) - 9.25
5. Brandon Phillips (40) - 10.0
The thing is, well, places don't shift very much in the 24 person and 16 person events. In fact, the first seven places remain the same.
Here's the results of the final 16 if you weight the scores:
Competitor Orig Place Orig Scr New Place New Scr
Holmberg, Graham (1) 48 (1) 71.67
Froning Jr., Rich (2) 51 (2) 74.13
Spealler, Chris (3) 69 (3) 88.13
Chan, Matt (4) 73 (4) 94.96
Salo, Mikko (5) 77 (5) 103.79
Malleolo, Austin (6) 84 (6) 112.29
Smith, Ben (8) 93 (7) 115.17 +1 rank
Burke, Patrick (7) 91 (8) 116.17 -1 rank
Phillips, Brandon (11) 121 (9) 133.96 +2 rank
Hackenbruck, Tommy (9) 111 (10) 133.96 -1 rank
Mackay, Chad (12) 123 (11) 138 +1 rank
Warren, Joey (13) 126 (12) 139.75 +1 rank
Egyed, Peter (10) 117 (13) 140.88 -3 rank
Orlando, Rob (15) 134 (14) 146.71 +1 rank
Khalipa, Jason (16) 136 (15) 151.33 +1 rank
Kelsey, Moe (14) 130 (16) 154 -2 rank
Similar results happened with last year's too. What this can mean is the way they score the game is not only simple, but holds well under cuts. Yeah, the best is let every one compete till the end, but it's not essential to get the best CrossFitter or World's fittest man.
PS: The reason why the places are shifting is that the guys at 7th to 16th are competing at a very close level to those near them. ANY changes in how scoring is done will likely shift them. I think how I did it is fair as it takes into account how the players performed against 45 men as well as the final 16.
Here's the points per place for the final 16 using my method if anyone cares.
1 : 6.33
2 : 6.83
3 : 9.71
4 : 10.67
5 : 10.71
6 : 13.13
7 : 13.21
8 : 13.58
9 : 13.79
10 : 14.63
11 : 15.58
12 : 16
13 : 16
14 : 16.67
15 : 17.58
16 : 18.04