How Much Faster Do Swimmers Get In College?

by SwimSwam Contributors 26

January 10th, 2017 College, Lifestyle

By Kevin Hallman

One of the most fundamental aspects of a swimmer’s season is improvement – only one swimmer can win Olympic gold in each event, but every swimmer has the opportunity to measure success or not against their own best times. As swimmers, we put hours of work into getting stronger, building more endurance, and becoming more efficient. Yet for all the choices people make about swimming in college, which team to swim for, what events to concentrate on, which division of swimming to compete in, most swimmers have very little idea how much or little these things actually have to do with their getting faster. Using a couple of years worth of data on college swimmers, I hope to shed some light on some of these questions.

Measuring improvement is tricky since so much of a season’s performance depends upon on choosing which events and meets to peak at. These meets and events can vary year to year. To account for this, I chose to compare a swimmer’s top three events to their top three events the following year, using the measure of overall percentile rank in those events. I then took any top times that were swum in events that both years and compared these to the previous year, trying to only count tapered-swim to tapered-swim improvement. Unfortunately I didn’t have reliable data for high school swimmers, so I could only look at how people improved from their freshman years onward: admittedly a large omission.

This chart should give an overall baseline for how swimmers improve over their college careers. Shown below is the average percentage improvement of all swimmers, broken down by division and gender.

Division Gender Avg. % Improvement
D1 Women 0.16
D2 Women 0.17
D3 Women 0.16
D1 Men 0.23
D2 Men 0.32
D3 Men 0.30

The first striking feature is that, on average, swimmers don’t improve much during their collegiate careers. To put a 0.3% improvement in perspective, a 50.00 second 100 yard freestyler who improved 0.3% would have swum a 49.85 the next season. The second is that men tend to improve about twice as much per year as women do.

The next graphic shows overall improvement data broken down by gender and year in school (well year of eligibility).

Year Gender Avg. % Improvement
Sophomore Women 0.27
Junior Women 0.04
Senior Women 0.14
Sophomore Men 0.49
Junior Men 0.16
Senior Men 0.11

It is clear from this graphic that swimmers tend to improve less as they age. Note that I don’t include freshman improvement. This is mostly because I don’t have access to high school times the same way as college, but also because the comparison would be difficult since high school swimmers tend to swim different events college swimmers do. I would be highly surprised if swimmers didn’t improve the most during their freshmen seasons.

Next, I dove into the data by breaking out improvement results by event. I’ll just show a couple of events here to illustrate trends. First, the data showed that breaststrokers tended to improve more across their college careers than swimmers who swam the other strokes.







Secondly, the data showed that distance swimmers improve less than sprinters. You can see hints of this trend in the 100 and 200 breaststroke plots, but it becomes much more clear when looking at the 100 and 1650 freestyle events.







Exactly why this is the case is a bit unclear to me. As a distance swimmer myself, I can acknowledge that part of the issue is that distance events take longer to get in shape for. Year-round commitments to swimming became more difficult the older I got as the rest of life got in the way. Another explanation might be that sprinting is more technical than distance swimming, which makes it possible to improve even if your strength and endurance do not. This might explain why breaststrokers tend to improve more than swimmers in the other strokes, since breaststroke is usually thought of as the most technically complex stroke. However, other endurance sports like running and cross country skiing see many athletes peak late in life, so this probably isn’t a complete explanation.

Another trend, though small, emerges across the board: fast swimmers tend to improve less than slower ones, even when looking at improvement as a percentage of the time. The data here doesn’t explain why, but there are a couple different reasons for this that I could postulate: better swimmers are closer to their training peaks, regression to the mean i.e. fast swims tend to be outliers, and the fact that drag forces increase proportionate to the square of velocity in water (approximately). But without some more data, it is difficult to tell.

Another hypothesis I wanted to test was to see how improvement factored into building the top college teams. Since the best collegiate programs tend to attract the best coaches and the most dedicated swimmers, I would expect them to improve the most. Improving your swimmers is one of the keys to building a great program, in theory. Of course recruiting and retaining good swimmers probably has a stronger correlation ultimately, but recruits are supposed to make those decisions largely on how much they think they might improve in a given program.

Shown here are the average improvement percentages for the top 25 men’s and women’s programs at last year’s NCAA Championship meets plotted against a metric that shows their team strength on the national level in 2016.









These graphs show that team strength and improvement are probably correlated, but not in a way that I would have expected. I’m not entirely sure if the fact that better men’s college teams improve their swimmers more while top women’s teams do not is really a trend or its just due to the smallish data set. I would tend to think the latter since I don’t really have a good explanation why it might be the case.

Getting faster in college is difficult; even the fastest swimmers at the top programs won’t improve much. So if you’re a college swimmer at least take solace in the fact that averages aren’t everything and that getting better than your competition may not be quite as challenging as it used to be.

Leave a Reply

Notify of
newest most voted
Inline Feedbacks
View all comments
4 years ago

D1 isn’t surprising, those athletes are already nearly at the pinnacle, take Conger his previous pr was a 4:13.87 honestly not a ton of room for improvement. Same with his backstrokes.

Reply to  Uberfan
4 years ago

somewhere there was a paper written by a graduate student on the theory of diminishing returns. Swimming was used as the “example” in this paper and it showed that the increased swim training in college programs actually provided very little return as performance (time) gains. If I remember right there was even an equation involved that took the swimmers time at age sixteen and a couple of other metrics and then “predicted” the athletes peak performance at age 20. Looked for it a little but couldn’t find it again. I remember reading the paper and wondering why college teams weren’t using the formula on all their recruits.

4 years ago

Any way you can make the charts on the bottom where we can zoom in? I can’t read some of the school names due to the low resolution.

Reply to  OldArmy
4 years ago

OldArmy – sure, I’ve asked our web producer to take a crack at it – should have a solution shortly.

4 years ago

This is great data! Three questions:
1. What is the source of the raw data for the summary analysis (Men/Women by Division, Year in College and Event) at the beginning of the article?
2. Are the charts at the bottom of the article that list the Division I men’s and women’s teams available for Division 2 and 3 teams?
3. Is data readily available to conduct the college-specific analysis at the bottom of the article over a three or four year period for Division 1, 2 and 3 colleges? This would create more data points for each college and yield a more comprehensive picture of how each college has performed over the past few years in total.

4 years ago

A key factor in swimming faster times in college is if the coaches develop ALL of the swimmers they recruit. During recruitment, coaches make many promises to young elite high school students about how they will use training strategy, racing strategy, and other state-of-the-art technical methods, etc., to help the recruits improve individual technique and swim faster. However, once recruits are on the college team, and due to NCAA rules TRAPPED in place, coaches can then ignore swimmers. There are D1 coaches who during practice, will not even watch many of their swimmers swim. Instead they focus on developing a few stars, and deliberately ignore the remaining highly talented swimmers who have untapped potential. These non-stars are used as filler… Read more »

4 years ago

I would have to completly disagree with this article and it’s findings. Several swimmers at NC State have been improving at alarming rates over the past several years. And much more than the .023% that are stated above. When can we get an article that talks about the improvements of swimmers at NC State? Im suprised how long pure talent has helped teams such as Cal and Texas to championships. Its only a matter of time until NC State pass them.

Reply to  WolfPack
4 years ago

If you reread the article, you would answer your own question. The 0.23% was the average across all men in D1. You can see in the plot at the bottom that NC State men are a large outlier in terms of improvement as compared to the other schools.

w0lFp4Ck Her0!11!
Reply to  Barry
4 years ago

dat wolfpack education

4 years ago

Having spent the last 8-10 years watching the kids that have come through our club program (and others) and then on to college, I have a hard time believing these numbers. I have seen many, many swimmers improve a lot more than .23%.

Plus almost all programs treat males the same way. The first thing they point them to when they arrive at campus as Freshmen is the weight room. Much of that first year is spent in building strength. And if you add 10 to 20 pounds of muscle to any well trained swimmer, I guarantee you he is going to be faster. And when you add that to the fact that many males are still growing in the… Read more »

Kevin Hallman
Reply to  SwimDad
4 years ago

Yeah I’m sure that missing the freshman data makes a huge difference since this is probably when people see the biggest improvement. One thing to note is that to get to the top of D1 swimming (or D2 or D3 for that matter) you will almost always have to have gotten a lot faster in college. The swimmers that stagnate usually won’t be the fastest.

I still remember seeing three Illinois state record holders all go off to Stanford: Danny Thomson, Burke Sims, and Connor Black. And watched as they remained fast, but didn’t really drop much time from their fastest high school events (500 free, 200/500 free, and 100 fly).

Also, while probably more common in D3 where I… Read more »

4 years ago

This is fantastic, but it’s only made me greedy for more information. Is it possible to get the results for more schools to help swimmers who are unlikely to be recruited by the top programs? However, using this analysis to identify programs with better coaching is tricky because it may merely reflect a willingness to take risks on recruits with untapped potential or even just high variance. For example, schools that show a lot of improvement may recruit swimmers who are new to year-round swimming. Even if these bets don’t pay off most of the time, the results could be skewed by survivorship bias; those swimmers who don’t improve are more likely to quit, and you will therefore measure a… Read more »

Kevin Hallman
Reply to  rich
4 years ago

On our website we have a bit more information. shows information for the programs and you can compare how they improve their swimmers and their attrition levels. I haven’t tested it, but it doesn’t seem that better improvement is correlated with higher attrition, but at least from personal experience there is certainly a survivorship bias.

I don’t know of anywhere that has information on how swimmers do once they leave high school, but would love to know if anyone else does!

Reply to  Kevin Hallman
4 years ago

Thank you for the link – it’s very interesting. If I am running the numbers right, the negative correlation between women’s team strength and improvement is much weaker outside of the top 25 teams. You are right that there doesn’t seem to be a strong correlation between improvement and attrition, though there does seem to be a positive correlation between attrition and team strength.

4 years ago

According to, Townley Haas went a 1:35 200 yard freestyle in Dec 2014, then his freshman year at UT went a 1:30 at NCAAs in March 2016. That is like a 5% improvement which is just unbelievable given how fast he already was.

Reply to  Mikeh
4 years ago

One could argue that it was actually much more than a 5% improvement. If you consider that “improving” to eliminate the whole 1:35 time period is not even remotely possible. I’m not sure of the physics, but there would be a theoretical maximum speed that a drag-limited human could possibly go given the limits of muscle power. Who knows what that would be for a 200 Free, but a reasonable guess might be something like 1:25.

So one could make the point that Townley dropping from 1:35 to 1:30 was more like a 50% improvement in terms of what is theoretically possible