TheRealTrain Course Comparison Database

For those of you who didn’t see it, here was an idea from one of our readers named Jiminy Cricket:

Etrain, how would you feel about creating some sort of comparison system for XC courses/meets? Since every course is different, it would be useful to be able to take someone's time and convert it into a time that works across the board (same idea as Speed Ratings, but this would be much more informal). I'm thinking we use the states course as the baseline/reference point.

For instance, from my experience, the updated Red White and Blue course (Schenley Park) is about 30-40 seconds quicker than the new Hershey course. Cooper's Lake is roughly 15-20 seconds faster than Hershey. The idea is that we get a bunch of people to comment with their own subjective conversions, agree on a consensus, and then have an idea of how fast an individual performance is on a known course (to most people). Hopefully, we cover just about every major meet throughout the season so we can judge each runner's time.

Again, just an idea. But I think this will be real useful, especially as XC starts back up and we don't know how credible breakout performances really are (flashing back to David Endres going like 15:45 in the 1st meet last year). Let me know what you think!

Personally, I think this is a very interesting idea and, with the right input from all of you, I think we could build our own little database of sorts. Hopefully you guys agree and we can work together to start piecing something together that will grow in the years that follow.

For starters, I’m putting together an excel sheet that will have some more numerical inputs to use to create a less subjective approach. Once I get the beginnings of that sheet up and running, I’ll post it up online and readers can add to it. I’ll explain all the ins and outs of the sheet once it’s ready.

In the meantime, I think it’s important to also have a subjective component of this. We need to make sure the numbers make sense with what we logically think because there are a lot of variables that go into this undertaking (time of year, weather, competition, natural variability, etc.). So I think we should start getting comments of course comparisons from the readers.

There are a lot of ways to contribute, but I think what is most important is to get a variety of ideas and discussions. Some ideas on how to share your thoughts:

1) Give a straightforward conversion. Maybe you think Lehigh’s 5k course is 45 seconds faster than Cooper’s Lake. Or maybe you think Carlisle’s course is 15 seconds faster than Big Spring’s. These give us a feel for how courses stack up and an easy way to compare times.

2) Rank courses in difficulty. Maybe you have run a bunch of different courses in your career. You can provide a quick ranking from easiest to hardest of the layouts. For example, say you raced 5 courses over the year. You may rank them from easiest to hardest as:

Lehigh
Carlisle
Belmont Plateau
Hershey
Salesianum

It doesn’t provide an exact # equivalence, but those are usually hard to hypothesize about. This gives an order that we should expect the times to trend. Remember, more than just hills goes into a course being hard. Bad footing, sharp turns or pavement sections could change your opinion even more than hills.

3) Provide some insight on courses that aren’t particularly well known. We know about the big courses, but what about some of the smaller invitationals? What about dual meet courses? If we get enough input, we can have data on these courses as well. People who are out in Erie County can benefit from those in the Catholic League and vice versa.

4) Provide info on how courses have changed over time. There are always changes going on to make something easier or harder. Maybe a course moved locations. These things can have an impact when we look back on year to year comparisons rather than just comparisons within a certain year.

Let me know what you guys think and be on the lookout for the spreadsheet as I continue to try and compile some data for us all.

Looking forward to hearing from you.

17 comments:

  1. Back in the day Etrains blog and/or the PA Don't Play blog had a lot of discussion about the legitimate length of some of the regular September invite course. The 5K times were ridiculous, many runners were hitting PR's in Sept that they'd never match in their career. A year after the discussions all of the sudden the times at those courses were slower...not because of weather or anything other than they were finally measured correctly as true 5K. These blogs were the driving force for getting wrong righted.

    ReplyDelete
  2. Lehigh to Hershey, +42 seconds. Rose Tree to Hershey, +24 seconds. Therefore, Lehigh to Rose Tree, +18 seconds. It's basic Algebra.

    ReplyDelete
  3. As a D3 guy, I did an unscientific sample of D3 guy's best times on Carlisle, Big Spring and Hershey.
    I took 8 guys (Affolder, Affolder, Henderson, Cupp, Shea, Kole, Giannascoli, Sullivan.
    Average times:
    Carlisle: 15:30
    Big Spring: 15:50
    Hershey: 16:15
    Pretty much as I thought. I am curious if there is any sampling of guys who ran Carlisle and Lehigh.

    ReplyDelete
    Replies
    1. William Tennent and Carlisle are both ridiculously fast, likely because they are either short or the courses are measured on the painted line, not the tangents the runners actually take when making corners. It's hard to take times from those 2 courses seriously. Based on my analysis, I'd add about 30 seconds from those courses to Big Spring, and add another 15-20 seconds from Big Spring to Hershey's states course. Pretty close to D3 guy's post above.

      Delete
  4. Schenley is tricky because there are so many variations of the course. I believe the Pittsburgh Central Catholic Meet (This meet is not run anymore) and the D8 meet run the course one way while the RWB Classic does a slightly different course. Additionally, RWB has changed its course multiple time over the past 3 years. The last 2 years have featured a hill at the beginning, but had 2 different finish lines. I know for a fact that the old course was not a 5k, despite advertising it that way. I wonder if the new course is he full distance, because the times have been slightly slower. In my opinion, the fastest course in the west isn't Schenley but is either Grove City or Sharpsville. I ran the same time on each course this year, but Sharpsville had Skolneckvich and Walsh, so it appeared to be the faster course. Coopers Lake isn't that fast. The fastest time, (and only sub 16 performance)last season was 15:55 by Mark Provenzo. Anyways, I can't help too much with conversion times, but here's my take on some of these courses from fastest to slowest.

    Sharpsville
    Grove City
    Schenley
    Cooperslake
    Brush Creek Park (Freedom invite and the MAC meet)
    Calu (Only 1 sub 16 ever recorded on this course)
    Gateway

    Wild card: Mingo
    Mingo host to a number of smaller meets last year but still featured some fast time from Conboy, Bumgarner, Pfeil, and Snodgrass. Though, I heard one of the coaches say, "I measured this course dozens times, its not a true 5k". It if ends up being the full distance, it would be faster than Coopers lake.

    ReplyDelete
    Replies
    1. Good points about Schenley. The first guy mentioned that RWB is 40 seconds faster than Hershey; that's not true anymore. Pre-2015, yes, maybe even closer to 1 minute cuz it wasn't a 5k course. Now, the finish line is a good 40 ft higher in elevation than the start, plus its a real 5k. I'll say the current layout is actually very close to Coopers in difficulty, maybe like 15 seconds faster than Hershey.

      Cal U (Tri States) is the hardest course I've ever run. I'd call it 5-10 seconds slower than Hershey.

      Delete
  5. Belmont to new Hershey is plus 15 seconds.

    ReplyDelete
  6. The PTXC at Kutztown is a fast course. That meet is absolutely stacked this year. Most top eastern half PA teams are there and Sam Affolder's new school from Virginia is showing up.

    It's tough to make a conversion because it's held in early September when most aren't in peak shape but maybe it's add 40 seconds to convert to Hershey in November.

    ReplyDelete
  7. fastest to slowest from my experience as a D1 guy:
    Tennent
    Lehigh
    Kutztown
    Council Rock/Tyler State Park
    Belmont
    Bowdin Park (NXR)
    New Hershey

    ReplyDelete
  8. I'd say that Tennent to Belmont is about 45-50 seconds faster. Belmont - Hershey is about 15-20 faster. An interesting one I've noted in kids I've coached, CR's Invite @ Tyler is about 5 seconds faster than what kids will run at the D12 meet at Belmont a month later. I like to use it as a measuring stick.

    ReplyDelete
  9. Lehigh
    Westtown (old Chesmont course)
    William Tennent (most likely short)
    Abington
    Unionville (New Chesmont course)
    Steel City (Coatesville)
    Belmont
    Hershey
    Salesianum (not even close to Hershey, absolutely brutal)

    ReplyDelete
  10. Tennent is wayyyy short. Tyler state park (council rock invitational) isn't that grueling of a course but can be tough to run a fast time on. Every one knows Lehigh. I think Cooper's Lake and Belmont Plateau are fairly comparable, maybe Belmont a hair slower. Hershey is probably the hardest in-state course (though I've never run Tri-States at Cal so I can't say). I actually can't speak on big spring, having never run there. Salesianum, in Delaware, is harder than any course in PA. Just an absolute monster of a course. In terms of running fastest to slowest, I see it as this, with Hershey being the baseline, and the amount of time I think the course runs quicker or slower than it.

    1. Tennent -65 seconds
    2. Carlisle -60 seconds
    3. Lehigh -45 seconds
    4. Tyler State Park -30 seconds
    5. Cooper's lake -15 seconds
    6. Belmont -10 seconds
    7. Hershey EVEN
    8. Salesianum +10 seconds

    ReplyDelete
  11. The Rose Tree course has the biggest hill of all the major PA courses. Hershey has more hills but nothing like that one.

    ReplyDelete
  12. Good idea for comparing all of the district/league courses: Take an average time of all of the runners that make it to states. You would have to make sure that weather and course conditions weren't drastically different between the meets, and also you would throw out a performance where an athlete clearly is not trying (Think Tony Russell helping his team win at 2013 Chesmonts). Over time the data would average out. What you do is assign a plus or minus score for each runner for each course, based on the performance at Hershey.

    So for example

    Runner A runs 16:00 at his league meet
    then he runs 15:30 at his district meet
    then he runs 15:45 at Hershey

    If you use Hershey as the benchmark for everything, his score for his league course would be +15 and his score for his district course would be -15

    What you would do is average all of the differences for each year and then look at the data over the years and see what the difference usually is, either taking a mean or median. Once you have enough data, you could consistently compare courses across the state. Since they are all being compared to Hershey.

    Also the reason you would only be able to use a runner that makes states is because most of those runners are all training to peak at states, and should therefore have similar improvements due to peaking which, would eliminate that as a source of error.

    ReplyDelete