Quote:
Originally Posted by plog
1. I think it has to do with what we are basing that .03 on. You did it against 1100, I did it against 1020 (the structure sheet says 3 percent of prize pool):
Vig% = [Total Fees] / [Total Cost]
[Total Fees] = 80 + 1020*.03 = 111 (Rounded from 110.6)
[Total Cost] = 1100
Vig% = 111/1100 = 10.1 (Rounded from 10.09)
2. There was only 1 $1100 tournament at the Wynn last year, but there are 2 this year: 1020 + 80 and 1025 + 75.
3. Using my formula for #1, I calculate a total of $106 in fees for the $1025 + 75, which becomes 106/1100 which is 9.6%.
4. I agree with you about the value of that comparison all together. I think I'm going to remove that comparison. Instead I think I am going to put every tournament into a cost range (<200, 201-500, 501-1000, etc.), take that range's average SPoints and Average Vig% and then determine what percent of each casinos tournaments in each range are above average SPoints and below average Vig%. Of course I'm open to other ideas and/ or range definitions.
You're right about the vig. I was using the wrong formula. As for the comparison methodology, here is what i think:
The idea of using buy-in ranges in order to categorize the tourneys makes sense, and I did the same thing last year when I tried to take your s-point work and come up with some methodology for comparing "value". But I can tell you (and I believe you know this), that was really hard to do in any sort of systematic and fair way. There is such a wide variety of "products" offered, even if one limits themselves to just NLH tourneys. There are turbos, survivors, satellites, bountys, MTTs, etc, etc. and even within these there are multi-day tourneys, and single day tourneys. Each of these variants has different characteristics which are attractive to different people. I think any truly rigorous analysis of casino offerings needs to take these things into account.
But perhaps there is a way to do this without introducing too many headaches. In addition to breaking down the buy ins into some number of ranges, break down the events into a "Length" category. I think three options would be sufficient. Something like "Turbo/1Day/Multiday" or however you would like to characterize it.
Turbos and most satellites would fit in the first category. You can even define these as having 20 minute levels or less. These are tourneys that are designed to complete within some number of hours and significantly less than a whole day.
All remaining satellites and single-day tourneys (the WSOP 235 deepstack, for instance) fit in the middle category. In essence, these are all remaining tourneys which do not bag up for the night.
And lastly there’s the rest which are designed to go multiple days.
Once tourneys are defined in this way, in addition to their buy-in range, I think making some sort of statistical analysis about average vig% and s-points would make sense and be useful as a year-to year comparison as well.
So, let’s say for example, you had 6 price ranges…. Something like the following, or whatever.
0-249, 250-499, 500-999, 1000-1999, 2000-4999, 5000-10001
You would end up with a potential of 18 rows of data for each casino. However, in reality, the number should be a fair bit less than this since not all venues offer tournaments across the whole price range spectrum or within all three "length" ranges.