Small changes to the course over the last few weeks have raised questions about the validity of the times run in August. What does the data say?
If you missed the forum discussion a few weeks ago, the folks who work hard to keep McAlpine in tip-top racing shape added a short segment to the second 1000 meters of the course. We'll go into much more detail in an article tomorrow, but the quick summary is this: over the years, the pathways on many sections of the legendary course have widened, which allowed athletes who run the tangents to get a shorter and shorter total distance. That process was accelerated by the construction at the park over the last year, and the end result was a course that was edging closer to 4900 meters, which is the unofficial threshold that we at MileSplit use to distinguish between 3 Miles and 5000 meters.
The current course is as close to 5K as you could ever want (closer than many other major courses in the state), and several authorities have declared that it is comparable to every period of the 35-year history at McAlpine Greenway. The one question that remains is the length of McAlpine before the current adjustment, which was the course for the Providence Invitational earlier this season.
We start by comparing this year's meet to the 2014 edition, using our Compare Meets feature (which you can access from any meet page if it has more than one year of results).
Providence Invitational 2015 vs 2014
It's difficult to compare from year to year without baseline data (we'll get to that soon). The weather could be wildly different, of course, but the bigger concern is that most runners improve over time. The questions then become twofold: what is the normal improvement from year to year at this meet, and did this year's meet significantly exceed that? The key there is the "Athletes Who Ran Both Meets" side of the comparison, so that we're measuring the progress of the same set of athletes from one year to the next, instead of comparing one group to a completely different group.
| Comparison | All Common Athletes | Middle 80% | Top 25% |
| 2013 to 2014 | -21.62 | -25.72 | -29.77 |
| 2012 to 2013 | -40.43 | -46.51 | -45.35 |
| 2011 to 2012 | -67.41 | -76.07 | -68.07 |
| 2010 to 2011 | -39.26 | -53.51 | -43.27 |
| 2009 to 2010 | -34.32 | -37.59 | -53.17 |
| 5-Year Avg | -40.61 | -47.88 | -47.93 |
| 4-Year Avg (excl. '13-'14) | -45.36 | -53.42 | -52.47 |
So, the improvement this year is clearly bigger, right? Well, yes and no. The jump in times from 2014 to 2015 is much higher than the average, but the 2014 times were much slower than average (note the much smaller improvement from 2013 to 2014). If you apply a roughly 20-second adjustment to correct for that, then the change from last year to this year is in the same range as it was in 2011-2012. That's at the upper end of the range we see over the previous 5 years, but it's not way outside the range. Also, it's worth noting that the weather was pretty close to optimal this year, which further exaggerates the difference between 2015 and 2014. If you look at 2-year intervals (2013-2015, 2012-2014, 2011-2013), which admittedly decreases the number of runners in the analysis, the improvements are actually quite consistent: -1:36, -1:17, and -1:36.
Results: inconclusive. This year's Providence Invitational times are on the bubble of being "too fast," but they're not clearly over that line. I suspect the course was in the 4915-4930 range for the meet in August, which is long enough to be considered 5K (although certainly on the short end of the range). Some supporting information for this assertion:
- The addition to the course (made after the Providence Invite) is roughly 75 meters on the shortest line, and the full length of the course is now 5000 meters, which would put the course at approximately 4925.
- Times from the Providence Invitational are not dominating the individual rankings for both genders. There are still several times in the top 25 of the boys' rankings, but on the girls' side the highest remaining time from the 2015 meet is #1 in the overall rankings. If the course was way short, it would have affected both genders equally; instead, it looks like the boys' race was just proportionally faster than the girls' race (which also happened at the adidas XC Challenge).
On the other hand, the times from the 2015 Providence Invite are still having a big effect on the team rankings. To be honest, I had hoped to compare them to the Wendy's Invitational, but now that has gone out the window. Although it's never too late to get something right if the information available is clear, this case is still pretty muddled.