Men's Individual Champion, Graham Holmberg, crosses the finish line in the Pyramid Double Helen event

2010 CrossFit Games Finals

Scoring the Games

Why the point per position scoring system is best.

The purpose of the CrossFit Games scoring system is to determine the Fittest on Earth. Before the Games, there was an extensive qualification process by which about 100 men and women earned the right to compete in the arena at the Home Depot Center. Just getting there established them among the world's elite of fitness. The Games just needed to differentiate among these elite.

After last year's Games, there was a lot of talk about proportional scoring. The great benefit of this system is that it rewards margins. If I beat you in Elizabeth by one second but you beat me in Diane by three minutes, we can say you're fitter because of the margins of victory even though we each won an event. This impact of marginal differences, however, becomes less significant as the number of events increases.

But the fatal flaw of proportional scoring is that the margins and proportions between different events are not equally valid indicators of fitness. Some workouts simply have greater margins, even as a percentage. In fact, the workouts with greater margins usually have specialized skills in them. In other words, when you dig in to the reality of proportional scoring, it favors the specialist by overly weighting workouts with special skills.

This year, the most common complaint about point per position scoring has been the impact of cuts. It is mathematically true that if you finish at the bottom of the heap before the cut, the cost is greater. If you finish 40th in an early event but still make it past the first cut, the worst your competition can do is 24th. (In other words, a bottom finish in the first four events leaves you with 40+ points, whereas after the cut, the most points you can get is 24 or 16. This makes deficits harder to eliminate as the competition goes on.) The complaint is that this means the early events are weighted more heavily.

This is not quite true, because there is a better way to describe this mathematical impact. It's not so much that it weights the early events more heavily, but rather it punishes glaring weaknesses. This system intentionally rewards athletes who finish closer to the top in every event.

This is all supported by the actual results. Neither Graham Holmberg, Rich Froning Jr or Mikko Salo finished any event below 16th. You could reorder the events however you want and their points wouldn't change. Chris Spealler had a 26th and a 22nd. If you capped the score on any workout at 16 (the fewest number of competitors in any event), he would have finished with 16 fewer points. But he was 18 points behind Rich, so the placings would have been identical. Even Austin Malleolo, who got 37 points in the Max Overhead event but still finished 6th overall, wouldn't have made the podium with a cap of 16 points since Speal would have benefitted from that also.

Going deeper, only three men in the top 16 after four events (right before the cut) had over 30 points on any event. On the women's side, no woman in the top 8 overall had over 20 points in any event, and no one in the top 16 after four events had over 30 points on any event.

What does this mean? It means that the scoring system works perfectly. At the end of the day, no one was taken out of contention because of some mathematical anomaly. The system rewards those athletes with the greatest work capacity across broad time and modal domains. It, like nature, favors the generalist and punishes the specialist with any glaring weakness. The athletes who finished on the podium at the Games are indeed the Fittest on Earth.

Rogue Fitness Again Faster Equipment CrossFit Journal

Proudly supporting the 2010 CrossFit Games

Photo Gallery

Camille Leblanc-Bazinett, 2nd overall in the first women's individual event Austin Malleolo, 6th overall, in the final event of the weekend
Follow crossfitgames on Twitter

Follow the CrossFit Games on Twitter

The CrossFit/USA Weightlifting Open

The CrossFit/USA Weightlifting Open

29 comments on this entry

1. Bryan wrote...

I don't think I can agree with the statement "You could reorder the events however you want and their points wouldn't change." From watching, it is apparent that Rich had one glaring weakness,a lack of ability to efficiently climb rope. You think that if rope climbs were in a workout before the first cut he would have done as well as he did? I've no doubt that he would have still done very impressive (the guy's a beast). And he would have done better at a rope climb earlier in the weekend as his body would not have already been punished by the eight workouts before. But I think it's inaccurate to say that it wouldn't have changed his position. I could have seen him easily picking up an additional 20-30 points if that workout had been workout number two.

I think we need to keep the point-per-place scoring system. However, once a cut is made, the previous workouts need to be re-scored as if those competitors who didn't make the cut were never in the workout. So once you cut down to 16, go back and re-score each event 1 through 16. This way all workouts are weighted equally. A glaring weakness in the first event is just as punishing as a glaring weakness in the last event.

The only way this method still isn't "fair" is this: If your glaring weakness comes up in the first workout, you're more likely to not "make the cut." But _if you do_ at least your glaring weakness is the first workout would be weighted the same as someone else's glaring weakness in the last workout. To _truly_ be fair, you need to keep the point-per-place system, and run the entire competition with no cuts. But that's impractical from a logistical standpoint.

All that being said, I still believe the people on the podium deserved to be there. Great job everyone!

2. Rob Silver wrote...

good argument, but try supporting it without hindsight bias

3. Rigo wrote...

Great points but there's still a pretty obviuos error. After the cuts an athlete can do horrible and only get a 16 where if the event happened earlier in the competition maybe the athlete gets a 42. Who knows what Graham would have ended up with in the sandbag event if he caompeted against 42 athletes instead of 24. Not to take anything away from Graham...he was awesome and deserves to be at the top.

It just seems to easy of a fix to ignore. If you cut from 45 to 24 make each place at 24 worth 1.875 and at 16 each place would be worth 2.8125. This means if you finish 16 out of 24 You would get a 30. It makes finishing at the top just as important in later events as earlier events. I hope the organizers really look into this and make some changes.

4. Kyle wrote...

I have no problem with this scoring system. As long as a system is made clear to the athletes before the competition, I don't think it is unfair. Every scoring system rewards some things more than others, I don't think there is a perfect system, and part of being a good competitor is understanding the scoring system and doing the best you can within that system.

That said, the argument in the second half of this post is misguided. No one should argue that the cuts problem would influence the final ordering every single year. If it didn't effect the results this year that is good, but hardly an argument that the system works perfectly or that the cut issue is not a real issue with this scoring system. In fact, we have a concrete example of where the cuts did have a large effect on the scoring: last year. Khalipa got 72 points on the run the first day. If that event had happened after a cut to 32 (or 24 or 16), his point total would have been the lowest (he was 28 points behind Mikko). So the system did punish a 'glaring weakness', but it would have punished it less if the workouts happened in a different order.

I'm not saying that Khalipa should have won 2009, or that the scoring system should be different. It seems fine to me, and Mikko won fair and square. But this system does reward those who don't have their worst workout early in the competition. I'm not sure why this post is trying to argue otherwise.

5. Greg wrote...

A lot of people are throwing around the idea that there shouldn't be cuts and everyone who qualified should be able to do all workouts. That would eliminate the different "weightings" of events pre/post cut.
It wouldn't necessarily take much more time to allow everyone to attempt each workout. There was a lot of time wasted by alternating between men & women (switching bars/weights, adjusting rings...).

Since the events were kept secret it made some sense to do it that way. Add in that there was only 1 video feed, and it makes more sense.

But, in future years, why not have men & women compete at the same time at different events/locations? If everything is getting archived and available for viewing just hours after completion it wouldn't be a big deal from a viewer's standpoint. Or, you could add another video feed and give the option of what to watch.

Announcing 2 events at a time for both sexes to complete rather than 1 wouldn't give anyone much more time to prepare. Men do one while Women do the other - then, once everyone has finished, they switch events.

6. sean wrote...

Persoanlly I think the ranking should be set by total time. Everythign can be wegithed by a time factor. Even when it comes to lifting weight. An example could be an instance of last years deadlift workout. Every lift subracts 5seconds of time from total time, stronger bigger atheletes would benefit as did faster atheletes on the run. Just saying form a spectators point of view. The most exciting events are those where you know what the deficit is and what you have to do to overcome it. The final with Speal and Khalipa is a perfect example. Everyone new what was at stake and what the times where to beat.

7. Leifur Geir Hafsteinsson wrote...

There's one minor error in the scoring system that I would like to see corrected for next year, and that's how the scoring of ties is handled. As it is, two competitors that are tied in an event, say in 3-4 th place, both get a score of 3.

It can be argued that this is incorrect, since a) it gives WODs with a lot of ties more weight than WOD's with no ties, b) it reduces the motivation for athletes to go beyond the current best score, since that won't affect their points.

The correct handling of ties is to give each athlete the average of the points their places indicate. In the example above, the total number of points for places 3 and 4 should be seven (3+4), divided between the two athletes giving each 3.5 points.

To make point a) more clearly, consider an extreme example: Let's say that out of 50 competitors in the max jerk event half lifted 300 pounds and half 200 pounds. With the current scoring the first 25 would have gotten 1 points each (as opposed to the more correct 13 each), and the remaining competitors 26 points (in stead of 38). The total points awarded for the event would be 675 points, in stead of the 1275 ( ((50+1)/2)*50) points each event should distribute between the 50 competitors.

So, in essence, the max jerk event would weigh twice what a wod with no ties, e.g., the double Helen, would.

Some might say that this is a minor issue that has no practical consequences for the scoring but that's not true. The mens deadlift event 2009 had this scenario happen (albeit less extreme), where well over 10 competitors each got 1 point for lifting the end weight successfully. This may well have affected who got cut and who didn't and created a scoring bias towards those with a strong deadlift.

The solution is simple. Give tied competitors the average points of the places they occupy.

PS. I very much agree with Kyle's second paragraph. The fact that things worked out well this year does not mean that the scoring system is perfect. What we have to accept that this is a sports event, we have to create rules (including scoring systems) and all systems are going to have pros and cons, especially when we face the enormous task of trying to identify the fittest persons on earth. The current system works well, it can be improved (as I'm pointing out above) and it can be applied differently (by rescoring after cuts as pointed out by Bryan), which would then result in a different event. I would welcome an analysis of how this would have affected cuts and results in the 2009 and 2010 Games.

8. Damon wrote...

clearly there needs to be a point penalty added on if you DNF a workout - it makes no sense that someone can DNF two WODs of the weekend and be called the second fittest woman on the planet when clearly women who ranked lower finished those WODs

9. Z wrote...

I like the points that are brought up here. Why is there a need here to assert that the "scoring system works perfectly" instead of simply stating that things happened to work out fine this year and CFHQ will continue to refine the scoring method just as CrossFitters refine standards of movements and techniques for performance?

The inability to admit even the most obvious or simplest of errors is a bit childish. You won't invalidate the Games if you say, "Oops, we f-ed up there, but we're working hard to address those mistakes." The author tacitly admits a mathematical impact exists to the cuts that were made, he/she simply chooses to ascribe outside meaning to them ("punishes glaring weaknesses").

If you want to punish glaring weaknesses early, might I suggest beginning with events that test general work capacity without testing (and penalizing for) high skill movements? For example, the sand bag event and the final three stage event would have been great work capacity events to put every athlete through. The more technical muscle-up/snatch event, DL/pistol/DU event could then have rounded out the competition after cuts had been made based on an athlete's more basic/generalize work capacity over several time and modal domains. Just a thought.

Regardless of some of these minor details, the event was an overall success and I want to thank HQ again for the live feed and the work they did to put this together.

10. Tony Budding wrote...

Damon,
This was made very clear before the competition started. The time caps are arbitrary for scheduling purposes. Every rep counted, so we could fairly rank everyone who competed. If you completed the workout before the time cap, we ranked according to time. If you didn't, we ranked by number of reps completed, below everyone who completed. There is nothing inherently different about finishing before the time cap.

11. Bryan wrote...

@Damon,
I think the penalty last year (DNF=disqualified) was applicable because they chose cut-offs that they believe a well-rounded elite crossfitter should be able to finish within. But this year they wanted to broadcast all heats of all workouts. To do this they picked short cut-off times knowing that not everyone would finish in that time frame and they had a scoring system in place (1 second penalty per rep not completed) to accommodate for this.
So in short, any cut-off time is going to be an arbitrary cut off. So you have two athletes, one that finishes 1 second before the cut-off, one that finishes 1 second after. Really, how much of a penalty should be incurred for those two seconds? Remember, the cut off is arbitrary, and has nothing to do with the fitness level of the athletes. In reality, these two athletes deserve to have only 1 point difference in their scores.
Cut-offs are used for scheduling, not for saying "if you can't finish in this time, you don't belong here." If you don't belong, that will show via your score, you don't need an additional penalty for not finishing within a time frame that is injudicious.

12. Damon wrote...

@Bryan - I understand your reasoning but if you can't finish a workout in even an arbitrary time that "a well-rounded elite crossfitter should be able to finish within" - then maybe you are really fit, fitter than 99% of the others on the planet, but you can't still be in consideration for "the fittest" when others did finish.

This goes to the heart of @Tony's response - the CF community expects those who design workouts to determine the "fittest on the planet" to also calculate within reason the amount of time for a cut-off for anyone below elite-status to finish. If 1 person came in under the 'arbitrary time cap' and 40+ other competitors didn't, then maybe there is an argument for the cap not mattering. But that wasn't the case here. Nor was it the case that athletes competed in so few events that their dominance in "broad time and modal domains" couldn't be tested. It was and you guys came up with AWESOME events this year to figure out who gets bragging rights. But when the majority of the pack finishes a WOD you can't say someone that didn't is as well-rounded as the rest of them. Fit? yes. An incredible athlete? Yes. Superior? No way. If you had them do Fran with a 5 minute cut-off (over double the most elite athletes would need to finish) and someone missed the 5-minute cap, it would be disingenuous to say that person overall is fitter than most of the rest of the pack that finished under 5. Especially if that person was ranked ahead of a Miko or Khalpia - fans would riot.

13. Damon wrote...

I'll just add that obviously you can't always make everyone happy and that overall the games were amazing and beat a lot of expectations

14. Steve wrote...

"Neither Graham Holmberg, Rich Froning Jr or Mikko Salo finished any event below 16th. You could reorder the events however you want and their points wouldn't change."

This is only true if you think that none of the eliminated athletes would have beaten them on the workouts after the cut. If the rope climb workout had been before the cuts, Froning might have gotten 30 points instead of 12. Same thing with Speal and the sandbag event.

Also consider the following: If the last cut had left only 8 competitors instead of 16, Froning wins the Games by 2 pts instead of losing by 3 (you can check my math on that, but I think it's correct).

I have no real problem with the "points per position" scoring system, but the timing and extent of the cuts have a real impact on the eventual winner. There are a number of ways to address this, most of which have already been discussed, but I think as it stands now, scoring methodology has too great an effect on the outcome of the Games.

15. Shaun wrote...

In the first two events there were a total of 1,035 points given out. In the last three events there were 136. This means that the last events were weighted less than the first. If, after each cut the subsequent events were scored so that each event gave out a total of 1,035 points, then the placings would have changed. Graham would still take 1st, Speal 2nd, and Malleolo 3rd. Froning would drop to fourth because he finished so low on the rope climb, and some people who were cut wouln't have been.

16. Danny wrote...

"fittest on earth"is just a title like "miss universe"', no one really believes it literally, nor is there a way to validate it.

17. Blaine wrote...

It seems like Bryan's scoring solution would work best in the future. As for the current system, it may have worked this year, but I doubt Jason thought it worked in '09.

18. Damon - Wasatch wrote...

@Danny. The way to validate it is to compete in the Games. As the pool of athletes gets bigger and Crossfit's definition of fitness continues to impact the mainstream there will be even more validity to this claim.

19. Chris wrote...

To me, this article looks like a HQ "spin-job" to avoid admitting the scoring isn't perfect. The statement that the current scoring method is "perfect" is ridiculous. I love crossfit and all, but one thing that is very annoying to me is HQs inability to admit when something went wrong. I hope HQ can put pride aside and change the scoring for next years games.

I think that Bryan have put some serious thought into Crossfit scoring methods and that his system would work.

Peace!

20. Vitaly wrote...

Rather than HQ beating its chest and saying "We're always right, our scoring system is perfect," and effectively blocking their ears like a child who doesn't want to hear any criticism, why not take on board the advice of the people who are just trying to make the competition as fair as possible?

It's obvious that the early events carry a far greater penalty for coming last than the later ones. How is that possibly fair? So many suggestions have been made for simple ways to evenly distribute the weight of each event over the duration of the whole competition, why ignore them? Also, Leifur made a great suggestion regarding the handling of ties that I'd never considered.

Specific suggestions aside, one question really bugs me. Why does HQ so vehemently reject constructive criticism (despite priding itself on being "open-source") from the very people that only want to see it thrive?

21. Damon wrote...

@Vitaly - exactly re: weighting of events. If someone doesn't finish a WOD and it was double-unders and they completed 5 reps, that can't possibly be the same as someone in a different WOD completing 5 rope climbs. Or as someone suggested, erase the points after each cut and start over. Why compare results for events in which your original competitors are no longer a part of? Or, if you’re going to give credits for reps completed then why would anyone who came in under the time limit stop working? Why couldn’t a first place WOD finisher do more reps to put a wider lead between her and those who didn’t finish?

22. Daigle wrote...

There are a myriad of arguments you can make for or against this system. There is clearly a mathematical "problem" with the system, as has been bludgeoned to death already. It weights early events very heavily due to the number of points available. In and of itself, that's not a problem, as everyone knows the rules beforehand. However, It places a much heavier weight on strength events as well (such as a 1RM deadlift, or clean, etc). The reason is generally there are a large number of ties, which creates a problem. Because of the unitary nature of lifts (i.e. many athletes clumping at say 225lbs, 275lbs, 315lbs, etc). Although the awarding of points FOR THAT EVENT is fair, meaning you're scored based on how many people you did better than in the event, when applied to the entire competition it creates an issue due to the number of available points.

Secondly, One of the issues created by the cuts and changing available points per event, is just creating place holders. Because there are only X number of points available in an event, it's possible that ALL other competitors are mathematically eliminated before the event even begins. This happened in the affiliate cup. Going into the final WOD, our affiliates team was in 3rd place (out of first by 18 points). There were only 16 points available, so the first place team could have finished dead last, the 3rd place team could have finished first and the outcome would have still been the same. It was IMPOSSIBLE for the team to win the competition (from 3rd place no less) before the workout even started. It's kind of anti-climactic when you go into a "Final Event" and it bears no impact on the final outcome, and 15 of the 16 competitors are basically just there for the WOD's sake without a chance of winning.

23. Spectator wrote...

What I'd love to see is a scoring system that, as does every playoff system, rewards those who have "made it through" the rigors of reaching the finals.

Once the desired number of athletes reaches the final series of events (if we are using the 2010 Games, that would be after the final cuts were made on Saturday evening), wipe out the scores. In the finals, everyone is assumed to have earned their place, and everyone begins at an equal footing.

This is, of course, similar to the playoff system in football, basketball, etc., in which the goal is do well enough in the season (or, for the CF Games, in the early WODs) so that you make the finals. Once you make the finals, well, it's anyone's game. Just as it doesn't matter whether you made it into the playoffs as a "wild card" or with an undefeated season, so too could individuals and teams have to prove themselves anew in the finals.

Of course, it is certainly true that all the same challenges and criticisms of a playoff system would be true here as well.

"What if you did well in the season [early WODs] but your weaknesses come up in the playoff?"
Short answer: that is tough luck.
My opinion: programming in the finals would need to be as well-rounded as the Games entire--that is, it would make some sense to have the finals consist of more than one scoring opportunity representing different time domains, etc.

"What if teams/athletes stop doing their best, and instead do 'just enough' to qualify for the playoffs [finals]?"
My opinion: there is no way to force any person or team to do anything, at any point in the competition. Everyone uses some sort of mental strategy, and some people/teams may choose to take their chances and not tax themselves early on. So be it--that is part of the adventure of sport and competition!

The bonus for spectators--and perhaps some competitors--is the satisfaction of knowing you've got full buy-in in the finals. In a playoff system, no athlete/team has the mathematical luxury of "coasting" or the psychological rationale of "I'm so far behind numerically, it doesn't matter": you have made it to the playoffs/finals, you deserve to be there, and every one of your competitors has no better and no worse statistical chance of moving any further.

What do others think? Do you think such a system would be desirable for CF? How could this be improved?

24. chris wrote...

I don't know if this have come up already, but a way to solve the problem of ties during a 1-5 RM event could be to use percentage of body weight as a criteria to separate the competitors. If 2 athletes makes a 500# deadlift, the one with the lowest body weight would be "stronger". Just a thought. There is probably some flaws in that system aswell.

25. Ollie_UK wrote...

Cutting places is fair enough and adds excitement to the competition but the place points need amending to avoid overly punishing early weaknesses. I think RIGO made an excellent point regarding the weighting of position points. This was also highlighted again by SHAUN. It would equally punish any weaknesses whether they appear before or after any cuts. It would also address the scenario mentioned by STEVE.
I think this is the main issue with regards to the scoring. As mentioned above, there will never be a 'perfect' system as every change will affect the athletes in one way or another.

In this sport it is impossibly hard to judge the true 'fittest' athletes. There has been great articles in the journal in the past about this. The area 'under the curve' is so complexed, it's not really 2D, it's more like 10D!!! But that's what makes Crossfit so amazing.

Scoring aside, best games EVER!!!

26. Hari wrote...

Would it be possible to marginally improve on the current system by using the same range of points on each WOD, even as the number of competitors decreases?

For example in the early WOD's, if there are 60 competitors, the points would be 1, 2, 3, . . . 60, as is currently the case.

If in a subsequent round, if there are only 24 competitors, the points would be 1.0, 2.5, 5.0, 7.5, . . . 60.

In the final round, if there are 12 competitors, the points would be 1, 5, 10, 15, . . . 60.

This might preserve the best of the current system while fully penalizing low performances in the final rounds.

27. Ben wrote...

One "problem" I have with the original post is this paragraph:

"The complaint is that this means the early events are weighted more heavily.

This is not quite true, because there is a better way to describe this mathematical impact. It's not so much that it weights the early events more heavily, but rather it punishes glaring weaknesses. This system intentionally rewards athletes who finish closer to the top in every event."

The fact is that the early events are ABSOLUTELY weighted more heavily. And the current scoring system may reward those that finish close to the top but it doesn't punish glaring weaknesses so much as it rewards muscle ups & squat snatches (wod 1) more than it rewards rope climbs & wall burpess (last wod). In another in last year's event the scoring system reward a 7k trail run more than it did a GHD situp. And while some of the top few competitors placements were unaffected, that is only by coincidence for this year. In some future year it may be the difference in winning or not winning.

In this particular year, if the events were equally weighted James Hobart, Spencer Hendel and Candice Ruiz advance to the Top 16. Unfortunately Jason Khalipa, Joey Warren and Carey Kepler would not have. Additionally, the order of the Top 16 that actually advanced under the current system would have been different in an equally weighted Games. This is the "revised" Top 16 for women:

Clever, Kristan
Thorisdottir, Annie
MacKenzie Voboril, Valerie
Pamanian, Jessica
Voigt, Rebecca
Foucher, Julie
Bergeron, Heather
Phillips, Christy
Leblanc-Bazinet, Camille
Kinney, Michelle
Smith, Lindsey
Umeda, Elyse
Brost, Cheryl
Plumey, Lauren
Kepler, Carey
Tovar, Stacie

And for men:

Holmberg, Graham
Froning Jr., Rich
Spealler, Chris
Malleolo, Austin
Salo, Mikko
Chan, Matt
Burke, Patrick
Smith, Ben
Hackenbruck, Tommy
Egyed, Peter
Kelsey, Moe
Mackay, Chad
Khalipa, Jason
Warren, Joey
Phillips, Brandon
Orlando, Rob

Again, in an equally weighted Games, Kepler, Khalipa and Warren wouldn't have even been in the Top 16. Austin Malleolo who place 6th and out of qualifying spot for next year's Games, would have in fact earned that spot in an equally weighted Games.

I think these things matter and a simple change to the scoring would address this. I laid out how I equalized the events in a previous post: http://games2010.crossfit.com/blog/2010/07/2010-scoring-system-announced,666/#comment-9313

28. Dean wrote...

Games were incredible, and HQ pulled off a great event for athletes and spectators alike, especially the final event. That being said a couple obvious issues.

If your adding a masters division I dont think programming should be materially different than that used for men and women. Rather maybe scaled.

I think with only 50 athletes there shouldnt be any cuts. Not sure its fair to athletes given aforementioned scoring anomolies. It would seem a simple adjustment that if you cut half the field, you adjust remaining field pts by half, thus recreating punitive nature of finishing last in an event.

RE DNF point attribution........... you almost had Annie win and she would have had 2 DNF events....hardly the outcome expected from the fittest. I think HQ needs to rethink their wod/time allocation. In the end, the fittest should be able to complete all the physical elements of the test.

Kudos to Tony and journal media for awesome coverage. It had non crossfitters totally enthralled and glued to the net.

Is there any limitation on the athletes in terms of medical treatments? I know you do drug tests, but if an athletes body fails ala Khalipa in 09, should he be allowed to continue if he receives IV hydration. Shouldnt the line be drawn that if you need that level of attention due to physical exhaustion, you effectively could not continue on your own physical resolve/muscle recovery, and should be DQ'd

29. Justin M wrote...

This article and the one it links were intelligently considered and well written. Both articles make some good points.

But, I think there are some issues still. I would love to bounce some ideas off whoever wrote the article.

A couple quick points:
1. Ben is right in that early events are worth more than late events. Rich Froning on the last event, Chris Spealler and Brandon Phillps on the Sandbag event all stand out. A really bad performance later in the competition doesn't hurt much because you can only lose so many spots. Someone wrote an article last year that was, "What if the Run were the last event?" In this scenario, Jason Khalipa would have been #1, and not #5. This year, it just so happened that the top athletes were a good bit ahead of everyone else at pretty much everything, and the athletes all tended to be pretty consistent. Any reasonable scoring system was going to place Graham and Kristin first. Next year might be different.

2. I don't really see how multiple events makes margin-of-victory less important. It just makes each event worth less and so one out-of-character event isn't going to change things much.

3. Consider my system: Proportional scoring based on the Median result that uses Standard Deviation rather than pure proportions and factors in Outliers so as to not greatly change the rest of a group. I made an Excel for this that I can e-mail.

4. You mentioned some events not being as good indicators of fitness as others. That's legitimate -- but just make them count less intentionally. I think it's hard to say that, last year, the Deadlift event (pre-cut) should count more than the Snatch event (post-cut).

5. There was something in the last article about some events being poor predictors of the winner. Something like a Shoulder Press is a single-modal event that clearly favors shorter, heavier competitors. Something like a 5k Run is a single-modal event that favors taller, heavier competitors. If you combine events that, in aggregate, even out, then they are quite fine. Obviously if you want just one event to give you a prediction of the winner, you are going to need a mixed modal event (i.e. the first event this year), but that doesn't mean it should be worth more...

6. I really dislike rank-based scoring, and by having it at the Games, it permeates to all the Regionals, and it is even worse there. Margin-of-victory should be something considered.

7. I also don't think we should "punish the specialist" so much with the scoring system. Using the SAT as an example... Someone that scores 600 on both Math and Verbal did not perform as well as someone that scored 800 on Math and 450 on Verbal. This should be reflected in the Scoring, imo. The Specialist will still do worse because of the increasing marginal cost of improving one single area of fitness, and the general declining marginal value of such an improvement.

-->

The 2010 CrossFit Games Official Sponsors

For more information on how to sponsor the CrossFit Games, see the sponsorship page.

  • PROGENEX
  • Under Armour
  • Rogue Fitness
  • Again Faster Equipment
  • MiR Pro Weighted Vest
  • Concept 2
  • WeightVest.com
  • Vibram Five Fingers
  • OMG*Omega3
  • Stud Bar
  • Regupol America
  • Mindbody Online
  • Forged Clothing
  • GarageGym
  • Gymboss Interval Timer
  • Inov-8
  • At Large Nutrition
  • LifeAsRx
  • LifeAsRx
  • Undefeated Sports Nutrition
  • SEAL FIT: Powered by NAVYSEALS.COM
  • Rage Fitness Supply: Leader in Functional Fitness Equipment
  • Stronger Faster Healthier
  • Watermans Applied Science
  • Rocktape
  • DeFeet
  • AquaHydrate