Hello all. Every once in a while, someone runs a report on the Pro Racquetball Stats website that exposes a bug. Data, coding, logic, etc. I thought i’d tell you about some of the bugs i’ve found and fixed lately, now that we’re getting a lot more attention, and hopefully highlight some reports you may not have known about in the first place.
If you were ever interested in the sausage making behind the page … then this is for you.
Here’s some bug-fixing logic done lately.
1. Miscellaneous Player Stats bug: this was exposed when people noticed that the site made it look like Kane Waselenchuk had never lost when winning the first game in a tourney. Turns out I was accidentally stripping the () parentheses from the scores (to do counts and averages) before doing the “lost the first game logic.” So now the report works. Kane is exactly 498-5 in his career (as of Portland 2018) when he wins the first game. So that’s pretty good.
To see the corrected report go here: http://rball.pro/27F927
2. The “Oldest player to” report was failing recently. As it turns out, I had a bad birthday for a player in my Player Profile data, which corrupted the age-calculation logic, which crashed the report. duh. I fixed the birthday, re-uploaded player profile data and it works. I call this the Ruben Gonzalez was amazing” report; he qualified for a main pro draw two months shy of his 60th birthday (!).
s1.proracquetballstats.com Report and Bug fix update.
Coincidentally … I now have more than 1,400 players in my Player Profile database, many with birthdays and other biographical information. I’m always looking for more data. If you’d like to update your data DM me. I also put up this form here that submits data automatically: http://www.proracquetballstats.com/…/player_profile_submit.… .
3.The Longest Game winning streak had an odd result in it. As it turns out, the loop logic for capturing game winning streaks was not properly accounting for qualifying CURRENT games won streaks. I noticed this while demoing the code; we were taking Kane’s current games won streak (which sits at 48 games as of the end of the Portland 2018 event) and “assigning” it incorrectly to the next player alphabetically in the player database. Whoops. So we fixed the logic to push in any active streaks properly. See the report now:
Kane is working on the 8th largest ever consecutive games won streak, which is now all the more difficult to increase because matches are best of 3 instead of best of 5. If the IRT had been playing games to 11 all this time, he’d likely be in the 64-straight game range right now.
4. Speaking of Winning streaks, did you know that Kane’s on somewhat of a large match winning streak right now? As of the end of the Portland event, he’s now won 70 straight on-the-court matches (as in, ignoring no-shows and complete forfeit losses where he did not take the court). Problem was, the report wasn’t showing it properly. The reason ended up being the way I coded his Sept 2018 loss in Laurel; the code wasn’t picking up the “wbf-ns” code I used. Fixed the code, and now he shows a current 70-match non-forfeit winning streak, good for 3rd best all time.
5. I fixed a small re-direction bug; when you show all the match results for an event, there’s a button that redirects you to the Quarters/Semis/Finals report for that season. Well, in the doubles side … my Q/S/F report can’t show the doubles results for that many players b/c it’d be way too wide. So now instead, if you click on that button you get a list of all the Doubles FINALS for that season. Here’s all the doubles finals from the 2017-18 season report as an example:
6. I fixed a couple of random data entry errors I’ve seen here and there. I prepare a spreadsheet and essentially “predict” the tournament from start to end and sometimes I forget to correct the winners and losers in the staging XLS, which results in data entry errors. I also find weird transcribing results from historical tourneys every once in a while and have to go back to the magazines to fix it. Today I had to pull up issues of Killshot from 1992 to fix some tourneys from that era.
7. Lastly, I added the season-ending points to the Season Seed Ranks report to provide some context for the players’ results from that season. This data is only really complete for later years where I have full rankings and point totals. The current ranking system went into effect prior to the 2002-3 season and the point totals have been pretty consistent since. Here’s a link to this report for the 2017-18 season with the new points field shown:
Thanks for being a fan of PRS!