I’m using WinTD as a comparison because I own a copy, and I don’t own SwissSys. I’m not actually comparing WinTD in my simulations, because I can’t. My simulations will take the same set of players, and run the same tournament hundreds of times, and compare the results statistically. To do that, I first have to implement my own interpretation of USCF standard pairings within my program, so I can use the simulation with that pairing method. To be confident that I have correctly implemented the USCF standard pairings, I want to make sure my program can duplicate WinTD results or, at least, that when I vary from WinTD results, I can be sure that the difference is due to picking a different, but still perfectly legal, set of pairings.
I think I’m at that point right now. My implementation gives the same results as WinTD in most cases, and when it doesn’t, I don’t think it’s any worse. However, I’m taking a break from that for just a bit. Right now my focus is on making sure I can deal with bye players and unrated players in every system that I simulate. I had been conveniently ignoring them in the simulations up until now.
Shortly after writing the last post, work got all topsy turvy and I had no time for leisurely programming. As a result, I wasn’t able to use my prototype program. However, it already has paid some benefit to me. Yesterday, I thought I had thrown the laptop into the car, but I had not. I didn’t discover, until it was too late, that I had no computer available to me, and would have the joy of doing pairings by hand.
As a result of the work I did on this program, I think I came close to getting the right pairings. I doubt I got all the color allocations absolutely correct, but I at least knew what I was doing and I think I came close. It also didn’t help that it was a day full of some really large upsets, making for some weird pairings. In the fifth round of a rated Swiss tournament, a player called my attention to the fact that there was a game scheduled that had a 1600 point rating difference between two players. I double checked it, and yes, that was correct. Very bizarre day.
Running pairings with nothing but pairing cards can be done, but I think it would take a bit more practice for me to feel comfortable with it. I’ll stick to the computer unless I absolutely must go the hand route.