crossfit games

Data Fit - Crossfit Open 17.2 Analysis

The second workout of the Crossfit open is officially over, and it was brutal...if you could see my palms right now, you would get an idea of the extent. Although the entire workout was tough, the real star of the 17.2 show was the muscle up. Looking at the analytics and the drops in numbers, it becomes apparent that the bar muscle up was a make it or break it point for the majority of athletes. 


If you're unfamiliar with the 17.2 workout we have listed it below: 

17.2 - Beware the Muscle-Ups
12 Minute AMRAP
2 Rounds of
50ft Walking Lunges (50/35)
16 Toes To Bar (T2B)
8 Dumbbell Power Cleans (50/35)
Followed By 2 Rounds of
50ft Walking Lunges
16 Bar Muscle Ups (BMU)
8 Dumbbell Power Cleans

This workout was crafted so that if you had Toes to Bar, you could do it RX (as prescribed) and then spend the rest of the time working on getting your first muscle up. The open has a certain amount of magic to it where people find the energy and motivation to do things that they have never done before and this year was no exception. We will show this later in the analysis. 


Our first analysis will be a comparison between completion numbers from 17.1 and 17.2. These can be seen in the table below.

Gender Type 17.1 17.2 % Diff
Male RX 180915 130105 32.6%
Scaled 34250
Female RX 138019 71768 63.1%
Scaled 56873

If you look at the table, you can see a significant drop off in the number of workouts completed RX (as prescribed). The numbers above start to tell the story of the effect muscle ups had on the completion rate of 17.2. 

Looking at this you can see a substantial number of people scaled this workout compared to 17.1. This makes sense, particularly given the difficulty of gymnastics involved. If you don't even have a chest to bar or are struggling with pull-ups, then it could make sense to scale it and get a better all around workout. 


If you're reading this article because you want to see where you stacked up against the competition, looking at the percentiles will give you a good idea. 

The graph below represents the overall distribution of scores in this workout for the RX division. Because the distribution is so heavily concentrated around the 78 rep range, we will look at a smaller range of results to understand the data better.

Number of Results (Y) vs. Total Number of Reps (X)

Number of Results (Y) vs. Total Number of Reps (X)

The biggest spike above represents the point at which muscle ups enter the workout. Around twenty-five thousand men and women did not make it past the BMU's. The other rather large bump is where the second round of BMU's come into play. The significance of percentiles and getting at least two BMU's can be observed in the table below.

Percentile Men Women
10% 78 55
20% 78 78
30% 79 78
40% 83 78
50% 89 78
60% 104 78
70% 114 80
80% 120 88
90% 136 114
95% 155 123

If you weren't convinced that the muscle ups were the real killer, the numbers above are very telling. When we first analyzed this data, we were wondering why so many seventy-eights showed up, but after further inspection, we realized that this was the exact number of reps it takes to get to your first muscle up. If you're a woman and can do two muscle ups, you jump from the twentieth percentile to the seventieth percentile. If you're a man who finished the first round of BMU's, you jump to around the fifty-sixtieth percentile.

The table below shows just how deadly those BMU's are. The one BMU stat is pretty cool though because it could represent the magic of competition. It is possible that over five thousand people completed their first muscle up this weekend.

Gender Bar Muscle Up Number %
Male No BMU 37521 29%
1 BMU 4027 3%
Female No BMU 47813 67%
1 BMU 1816 3%


In the spirit of competition, we thought it would be fun to break some of the stats down on a region to region basis. Since the muscle ups were the defining factor in the competition, we decided to break out each region and put the percentage of male and female athletes who were able to complete one or more muscle ups on the map.

For those of you who might not be able to see the results above, we have some more data in the tables below.   

Female Results

Region Results Rx Scaled 95% 50%
Africa 3091 1675 1416 115 78
Asia 2592 1368 1224 92 78
Australia 8950 5549 3401 122 78
Canada East 5346 3046 2300 123 78
Canada West 3079 1770 1309 117 78
Central East 8091 4569 3522 123 78
Europe 18360 10839 7521 123 78
Latin America 9324 4058 5266 120 78
Mid Atlantic 9182 5212 3970 128 78
North Central 9747 5230 4517 126 78
North East 11461 6095 5366 122 78
NorCal 4105 2393 1712 121 78
North West 5276 2970 2306 122 78
South Central 9808 5186 4622 122 78
South East 8866 4912 3954 126 78
SoCal 4779 2875 1904 125 78
South West 6584 4021 2563 126 78

Male Results

Region Results Rx Scaled 95% 50%
Africa 5145 3961 1184 152 88
Asia 5060 3889 1171 139 85
Australia 10175 8553 1622 157 89
Canada East 5744 4680 1064 160 89
Canada West 3022 2460 562 149 87
Central East 9144 7347 1797 159 89
Europe 32103 25482 6621 158 90
Latin America 15906 11387 4519 144 92
Mid Atlantic 10257 8090 2167 157 88
North Central 11095 8763 2332 161 91
North East 12389 9570 2819 157 88
NorCal 4652 3752 900 162 90
North West 5166 4158 1008 158 89
South Central 10735 8513 2222 159 89
South East 10724 8660 2064 157 90
SoCal 6061 5016 1045 162 90
South West 6977 5824 1153 159 90

It's interesting to look at the ninety-fifth percentile (95%) and see how that number stacks up on a per region basis. Many people like to argue that the regions are proportioned poorly, but we can see that in general region size was not a good predictive factor in who had the hardest region (as based on the 95%, not the top 20 performances). Top performing regions for the men were Northern (162) and Southern California (162), while top performers for the women were the Mid-Atlantic (128), with three other regions coming in second.

Lastly, if you enjoyed this article and are looking to perform better in next year's regionals, make sure to download our FREE app on iOS or Android.

Data Fit - Crossfit Open 17.1 Result Analysis

We are heading into 17.2 and I know the first thing you want to know after you complete your workout is where you've placed amongst the competition. The leaderboard is an addictive source of live data that you can examine and explore. You can even find out facts like where you've placed compared to people with your name...I placed fourth among other Andrews in my region in 17.1. Whether or not you are competing, it is always fun to see how you are performing relative to the other athletes. So I thought it would be cool to show some of the data I was able to pull from the 17.1 workout!

Side note: If you find the Crossfit leaderboard crashing this weekend we've also created an open source leaderboard that should provide more stability. You can also search for results by affiliate!

2017 Crossfit Open Leaderboard


The 17.1 workout was fairly simple so a lot of people were able to complete it RX (as prescribed). Although the workout was simple, it definitely favored Big Engines.

We analyzed over 350k results over the entire CrossFit leaderboard to bring you the most accurate results. These are the numbers of results, compared with last year.


Gender Year Total Results RX % Growth
Male 2016 159386
2017 211254 180915 13.5%
Female 2016 120117
2017 157989 138019 14.9%

If you look at the numbers, it is pretty cool to see the dramatic increase in results posted. I am excited to see if the results posted from 17.2 hold up to the increase in athletes competing shown above. I say this because there could be some factors that came into play in the numbers above. For example, 16.1 was much more challenging skill wise (chest to bars) but 17.1 wss much more challenging mentally. I think the muscle ups in 17.2 will give us a good perspective. 

If you're interested in comparing your results to the rest of the field I included some interesting percentiles below. For the rest of the analysis, every rep below 225 (the total in the workout) was counted as +1s to your time.

Percentile   Male Time (Seconds)     Female Time (Seconds)  
10% 1267 1270
20% 1248 1248
30% 1232 1229
40% 1216 1213
50% 1207 1204
60% 1178 1163
70% 1121 1106
80% 1049 1039
90% 953 955
95% 882 890
99% 765 785

The first number that really stands out to me was that more than fifty percent of the competitors did not finish under the time cap. If you beat the time cap, congratulations you are in the top 50 percent of athletes. But if you didn't beat the time cap, don't be too hard on yourself. The time cap was hard to beat and it doesn't mean you aren't fit.

Although the percentiles are cool to look at, the graph below tells an even more interesting story! The gap in numbers between men and women were so close that I just combined the data into one graph.

Number of Results (Y - Axis) vs. Time (seconds). Men are Blue, Women are Red

Number of Results (Y - Axis) vs. Time (seconds). Men are Blue, Women are Red

We will improve this analysis as the workouts continue, and include data from 17.2 and onwards.

Remember to check out our app 38Plank if you want to create, share, and compete with your friends. If you like the CrossFit open, you should see what other workouts people have created. Just search on the #crossfit.

-- Andrew Cole


If you enjoyed this article and are looking to perform better in next year's regionals, make sure to download our FREE app on iOS or Android.

Data Fit - Crossfit Open 17.4 Analysis: How Much Did the Community Improve?

17.4 is.... 16.4

Every year during the open one of the workouts is a repeat of a workout from a past open. This gives everyone in the chance to measure their performance not by how they stack up in the community, but by how they have improved themselves. Personally, I think this is a healthier form of competition, as it lets you step back and reflect on your year of training to answer the question:

Did I improve my fitness since last year?

In our first three 38Plank Crossfit Open Articles, we included some fun data points, graphs, maps, and diagrams analyzing competition and performance between regions and the community at large. Although these are all really fun to look at, we decided to focus more on individual improvement and highlight how far the Crossfit community has come in the last year as a whole.  At 38Plank, we believe that you are your biggest is the main inspiration for our app after all! Don't get us wrong, we love some healthy competition, but fitness should ultimately be about challenging yourself and working to get better every day. 

So, in order to honor this belief, we decided to also include some fun interviews with a diverse group of athletes. We hope that their stories also help inspire and motivate you in your training :)

Improving Fitness

For the first part of our analysis, we wanted to see whether or not the Crossfit community improved as a whole since last year. To keep it short, we KILLED IT! 

"The average improvement of the entire CrossFit community who did RX in both 2016 and 2017 was 9.3 reps, or roughly 5.7%"

This number is from 90k+ results of people from different fitness backgrounds, cultures, and geographic locations, making a strong case that if you are looking for a new workout program that will improve your strength and overall fitness, Crossfit could be just the one for you! This is also probably the largest ever study of a fitness program or methodology ever conducted.

Let's break the community numbers down by male / female and also separated by region to see who had the biggest improvements.

Note: RX16 & RX17 means the number of people who posted an RX score in both 2016 and 2017.

Men's Regional Data
Region Reps Gained Percent Gain RX16 && RX17 Scaled to RX
Africa 13.18 8.04% 1501 150
Asia 13.31 8.39% 1272 161
Australia 8.20 4.81% 3754 241
Canada Eat 8.94 5.13% 2277 155
Canada West 8.40 4.93% 1231 88
Central East 7.77 4.53% 3740 243
Europe 11.36 6.80% 8967 795
Latin America 14.59 9.20% 3846 510
Mid Atlantic 7.52 4.36% 4087 254
North Central 7.58 4.37% 4407 307
North East 7.37 4.32% 5026 361
Northern Cali 6.61 3.85% 1960 91
North West 6.58 3.80% 2121 142
South Central 6.78 3.96% 4261 239
South East 6.92 3.99% 4176 227
Southern Cali 6.73 3.92% 2554 136
South West 6.98 4.09% 1925 155
Womens Regional Data
Region Reps Gained Percent Gain RX16 && RX17 Scaled to RX
Africa 14.98 9.88% 611 163
Asia 21.31 15.41% 471 134
Australia 9.98 6.26% 2465 462
Canada Eat 10.22 6.24% 1427 278
Canada West 9.23 5.67% 848 171
Central East 9.09 5.63% 2251 408
Europe 13.60 8.58% 3616 843
Latin America 16.38 11.05% 1328 466
Mid Atlantic 8.99 5.59% 2591 497
North Central 9.73 6.00% 2486 509
North East 8.88 5.51% 3014 655
Northern Cali 7.91 4.90% 1268 208
North West 8.14 4.95% 1457 259
South Central 9.56 6.09% 2508 471
South East 9.22 5.74% 2308 388
Southern Cali 10.20 6.32% 1500 254
South West 8.99 5.67% 1925 298

Looks like Latin America, Asia, and Africa are improving at a rapid pace. It's great to see the normally underrepresnted regions improving fast and likely catching up to everyone else.

All Reps are not Equal

We see that the entire community improved by ~9 reps, but what type of movements were they? 9 handstand pushups and 9 wall balls are not created equal, so we need to figure out what percentage of both men and women made it to each round to be able to base our information.

First lets look at the overall scores in the community.

Red is Men, Blue is Women

Red is Men, Blue is Women


As you can see a large group of the population was cut off at some point during the handstand pushups. Using this information we can better understand our results from before. On average, a large proportion of those extra 9 reps were from the handstand pushups. Looking at the table below, we see the breakdown percentages of the community. 

Percentile Male Reps Female Reps
10% 123 110
20% 149 135
30% 165 151
40% 166 164
50% 171 165
60% 176 166
70% 181 172
80% 189 179
90% 201 191
95% 215 204
99% 254 245

We see that 72% of the men, and 42% of the women who did the workout RX were able to get one Handstand pushup. That means a majority of those 9 rep improvements were done on this portion. 

Scaled to RX

For most of our analysis, we have focused on the RX community, as that provides us with the most consistent source of data. That may make the data science easier but Since the addition of the scaled category in 2015, many people have been going scaled, increasing the number of people who can compete in the open. In this workout, many people scaled for one of two reasons, either the deadlift was too heavy, or they are not able to get a handstand pushup. But anyone who improved their fitness since last year may have shored up these gaps and moved from scaled to rx, which in its own right shows an improvement in not only fitness, but confidence. The question we can ask ourselves then is, "How many people went from RX to Scaled" and use that as an number to show community improvement. 

The number of people who went from scaled in 2016 to RX in 2017 was 10,719

Thats a massive number. Deadlifting 225/155lbs x55 is no easy feat, and for an athlete to improve enough in one year to go from not being able or confident enough to tackle this, to going RX in 2017 shows an incredible amount of improvement.

Individual Stories

Looking into the data of the Crossfit Open is incredibly fun (at least for me) and insightful. It helps us as athletes put our scores into perspective and evaluate ourselves on a larger scale. But when you dig down into it the data does not tell the whole story, and doesn't answer the most important question that anyone can ask, "Why are we doing this?".

To find the answer to that all encompassing question, we interviewed crossfitters that improved and asked them questions about how and why the worked through out the year to improve. These stories are incredible testaments to the power of dedication, hardwork, and motivation. I am so happy that the athletes let us share their deeply personal stories why.