Jai-alai: It’s All in the Stats
By Steve “Straymar” Martin
Back in the fall of 2018 I presented my programmer with several challenges, one of which was creating an automated database for Dania Jai-alai results in Excel. I knew he was an Excel whiz but never dreamed I would end up with such a powerful tool with amazing capabilities! I had worked with this programmer for a number of years and was well aware of his Excel abilities as he had been called upon by the company where we both worked to perform some monumental Excel tasks and he always came out shining in the end and was known as the Excel guru. He was the superstar of Excel, the rising star and definitely going places.
The task took a lot of upfront work, which also included familiarizing my programmer friend with the basics of Jai-alai, all while he was working his full-time engineering job, along with putting in all that unpaid overtime every week that’s expected of engineering professionals, yes FREE overtime, imagine that! Having been an engineer myself for many years, I was well aware of that concept at several companies where I worked, LOL, and understood the challenges and that this special project request certainly wasn’t going to happen overnight.
One of the major concerns in this project was how to deal with scratches, and that’s was fairly quickly resolved, being the one area where the automated data base required some manual updating after up-loading the entries. Also, to ensure data quality, since the rule of thumb on data is “Garbage in Garbage out” I had to maintain the integrity of the data by keeping on top of not only the early scratches, but also any late changes that can occur during a performance.
Now understand, that when I went to college nobody had a personal computer, there was no such thing. In fact when I was a senior in high school, calculators first came out and were costing around $100 each. That was a helluva lot of dough considering this teenager was getting paid about $2.00 an hour for minimum wage jobs or $3 for mowing someone’s yard. Yes, imagine that, teenagers actually mowing lawns, that’s something one rarely sees anymore, sad but true.
On my college campus there was one computer, a very big computer. Sure, we had a programming class my freshman year where we learned to program in Fortran. We would write our programs, then had to go to a computer lab room near the campus’s only computer where there were a group of terminals where one would create a punch card for each line of code in our computer program. That resulting stack of punch cards, all in order for the sequence of the code for the computer program, would then be handed to a person at the door of the computer room , and your computer program job would be put into a que. This meant you had to come back the next day to pick-up the wide computer paper output, and your stack of punch cards, hoping you hadn’t made a single mistake in creating your punch cards, not a single typo or programming code error. A mistaken comma for a period would mean you have to repunch that card, replace it in the correct location in your stack of punch-cards, and re-submit your job to be picked up the following morning. Amazing how far we’ve come since those days! Now, my programming friend is a lot younger than me, in fact, I’m more than old enough to be his father, so he was well versed in the newer methods and programming techniques, especially from a top engineering school. Plus I had remembered from days of working nearby his desk at work where he once mentioned that he loves a programming challenge. So when I had initially contacted my friend and former co-worker on this project, I first mentioned “I’ve got a programming challenge for you.” Sure enough, that one statement was taken as a challenge and he never looked back after that.
Fortunately I had kept good records for scratches for all performances, so when we went back in history to capture all of the Dania fall statistics back to September 15, 2018, I knew my data was pretty accurate. I would always print-out the Dania entries sheet and kept a record of scratches, including other hard to get info such as final pool sizes for some betting pools. Pool sizes have always been important to me since the late 1970’s, especially if you want to test a new system, that’s another topic I won’t expand on at this time.
The end product was a pretty amazing database and using Pivot Tables in Excel with this database is a really powerful tool. It allows one to put the data in a useful format to make comparisons and you can find interesting trends and sort data quickly in many ways. I had lots of experience with Excel and Pivot Tables but no experience with the behind-the-scenes Visual Basic programming. Now understand my personal database does not collect payoff information; that was never my intention from the start. I didn’t want the programmer to waste precious time on payoff history with all the refunds and consolation payoffs that come into play with Jai-alai these days. The intention was to use the database to assess and compare player’s abilities for use in a Jai-alai game computer simulation.
The same database methodology was employed in the first half of 2019 to create a database for Magic City, however, data for at least the first half of the 2018 inaugural season was not included as players were new to Jai-alai let alone competitive Jai-alai games in front of fans. They were however, highly trained athletes in other sports, but not exposed to Jai-alai since childhood like the typical Jai-alai players at Dania. The famous Joey Cornblit is a different story though and quite fascinating since he started playing at age 12 and turned pro at age 16. The Magic City Jai-alai experiment involved taking athletes unfamiliar with the game and training them to play the game over a short time period, a very interesting experiment indeed, and it turned out quite successful too. Players at Magic City all progressed differently on the learning curve, with some learning much faster than others, and that shouldn’t be any surprise. So to include data from early in the 2018 season would have corrupted data quality for generating accurate player assessments to use at the beginning of the 2019 Magic City meet in July.
Here is some data you’ve never before seen from the two databases for the top singles players from Dania and Magic City Jai-alai:
Erik at Dania – for all 7-point singles games from 9/15/2018 through 3/15/2020:
GP-W-P-S record was 438-115-79-61 for 26.3% wins and 58.2% ITM
Erik – same data group but only the difficult posts 5/6/7
GP-Wins is 177-35 for 19.9% wins, was 48.6% ITM from those posts
Now when Erik had the 8-post for 48 games of the 438 games, he had 15 wins and 0 places, yes 0, and had 7-shows. Sounds like another story to be told there about post 8 and running the game out, and if stopping at 5-points, that 5-points is not good enough for a place in many instances, just thinking off the top of my head. It would be interesting to know what percent of occurrences Erik won game point for a perfect game from post 8. I would imagine it’s a very high percent?
Douglas at Magic City – for all 7-point singles games for his entire career to date from 7/24/2019 through 5/31/2020, note they were all late-game singles:
GP-W-P-S record was 344-132-47-41 for 38.4% wins and 64.0% ITM
Douglas won 3 of 4 games the first performance of his competitive career and won his first two games played. It’s fascinating that he picked up a cesta for the very first time in the first quarter of 2018, initially just for physical therapy from shoulder surgery in December of 2017. This Magic City phenom is only 20 years old and turns 21 later this month. As Stu has stated, Douglas is a natural in two areas; court positioning and catching.
Douglas – same data group but only the difficult posts 5/6/7
GP-Wins is 182-66 for 36.3% wins, was 61.0% ITM from those posts
Note that Erik had full benefit of random post draw for all of his games while Douglas did not. Erik played 40.4% of his games from posts 567 while Douglas had those difficult posts 52.7% of his games.
Now the full story at Magic City is that Douglas was so dominating in singles games, the random-post-draw ended for Douglas for a brief period from 8/28 through 10/6/2019. Note that Magic City Jai-alai took a week off during the Magic City casino 10-year anniversary celebration and when Jai-alai returned on Oct-16, new players Ikeda and RonRon were on the roster and started playing. During 8/28 to 10/6/2019 Douglas was not placed in posts 1 or 2, and only received the 3-post once (won that game).
Here are the Douglas 7-point singles game statistics during that period where random-post-draw wasn’t used for assigning his player posts:
Post-GP-W-P-S
3-1-1-0-0
4-22-10-0-4
5-51-16-5-5
6-44-20-4-5
7-41-15-4-4
8-12-6-1-1
Overall during this period, Win percent for Douglas was an astonishing 39.8%, even better than his overall career average of 38.4%. Apparently the post changes did not deter Douglas. His lowest win percent during this period was 31.4% from post 5 and best was 50.0% from post 8 not counting the 1-start with 1-win from post 3.
Jai-alai databases can be very useful tools but require time to maintain them. I find them to be both interesting and informative. Anyone tackling a project like this would employ their own techniques in their own way I’m sure, but the bottom line on Jai-alai is “It’s all in the Stats.” The stats show the facts and the facts paint a picture and there are many pictures to be seen if you have the right tools and know how to use them. These snapshots of Jai-alai stats presented here are just a tiny fraction of the wealth of information to be gleaned from the data.
I’m sure that Jeff “Laca” Conway’s new website will be a refreshing change and outlook on the exciting sport of professional and amateur Jai-alai today and I wish Jeff the best of luck with this new website (pelotapress.com). I’m honored to be asked to contribute and also be sure to check on the other new website named “SayHiLi.com” for more information on interesting statistics, those important early scratches and scratch/injury history, where I also contribute regularly.
