Welcome to PokerForums.org

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Results 1 to 6 of 6
  1. #1
    Private
    Join Date
    Nov 2014
    Posts
    4
    PFO Points
    40

    Default Variance or Is it!

    Hi

    First post so be gentle

    The subject of variance in poker is driving me nuts. Not the variance itself, but trying to understand whether or not I am a profitable small stakes mtt player or not. Recently I started a 1000 game experiment to see what my ROI would be after playing those 1000 mtts. As usual things got off to a flying start with me being $600 up (should have said only playing relatively small buy-ins ABI $7 ). Currently after tournament number 670 I am now $800 down!!

    I know that we are not yet at 1000 and I understand that 1000 might be considered too small a sample too but what I really want to be sure of is how you differentiate between variance and / or the fact that your game is just not up to it.
    To illustrate how daft this all seems to me. Given my current state of down $800 and ROI of something like -17% all seems hopeless but if tonight when I sit down to grind I win (not necessarily likely but certainly not impossible either) say a $3.30 R/A on stars for $3000 then my results would suddenly do a cartwheel and my ROI would go into an unrealistic positive state and my bankroll would be significantly up.

    This is all so woolly to me and surely there must be a better way of deciding if you are a winning player or not.

    I would really welcome anyone's constructive comments or thoughts on this or any articles you can point me too because it is driving me a bit nuts (well even more nuts than I already am) trying to understand it all.

    Thanks

    Mike (UK)

  2. #2
    Super Moderator Xopods's Avatar
    Join Date
    Jan 2009
    Posts
    4,281
    PFO Points
    34646

    Default

    The nature of huge online MTTs with thousands of players is such that it's essentially impossible to establish a firm ROI.

    Let's say the tournaments you're playing have an average of 2000 entrants. On 1000 trials, then, an average player should make the final table about 4-5 times. How you do overall has both to do with how often you make the final table and how well you tend to do when you get there, and the difference between 9th and 1st is generally a factor of at least 10, often more depending on the exact size and structure.

    So, you could easily have a bad player who only makes the FT twice, but luckboxes 1st and 3rd on those two attempts, and another guy who gets there six or seven times but runs bad at those final tables and never manages to do better than 4th... the first guy's going to have a higher ROI than the second, even though he's probably theoretically -ROI in the very long run.

    I've now got an overall 50% ROI on over 3000 games. But I was probably about 200% ROI on the first 1000 and -50% on the latest 1000. Part of that is variance, part of it is tilt and life cirumstances... but it goes to show that even 1000 games is not nearly enough to get even an approximate idea of a player's quality.
    Quote Originally Posted by pikachar
    Honestly didn t read OP I just got to say xopods is right on the money
    I design board & card games!: http://www.benefactum.ca

  3. #3
    Private
    Join Date
    Nov 2014
    Posts
    4
    PFO Points
    40

    Default

    Hi Xopods

    Thanks for that input. Very useful. I especially like the bit about it being virtually impossible to establish a firm ROI. All good food for thought.

  4. #4
    Private
    Join Date
    Nov 2014
    Posts
    4
    PFO Points
    40

    Default

    I have another related question: If we assume that a fictitious player is a proven profitable player over a large sample as our starting point. If he/she was then to go on a very long losing run. How many tournaments could they go without a significant cash ie final table or first 3 places before recovering again.

  5. #5
    Super Moderator Xopods's Avatar
    Join Date
    Jan 2009
    Posts
    4,281
    PFO Points
    34646

    Default

    That depends first and foremost on the size of the tournaments, secondly on their structure, and thirdly on the player's style.

    In the extreme case that you're playing 4000+ player Turbos like the Hot $11, the average player is only making final table about one time in 400, so even a winning player could easily go that many games without a final table.

    Variance is higher in faster formats because edges are smaller and there are so many preflop all-in races. So a turbo player could go much longer than a non-turbo player between final tables, for the same field sizes. Conversely, multi-day live events are fairly low variance because of their slow structure, which is why you see the same guys showing up at final tables relatively regularly.

    Finally, different players emphasize winning, final tabling and cashing to different degrees. Someone who tries to have a high ITM will make final table less often, while someone who concentrates only on winning and ignores bubble concerns will final table more often but have a lower ITM.

    Looking at my own stats, for scheduled multi-table tournaments with 181+ players, I've played 2047 games, most of them turbos, with an average field size of 2165, and have 12 top-3 finishes (5 first, 4 second, 3 third) and most of those were in fields below that average. My five wins, for instance, were in fields of 1265 (Hot $22), 470 ($2.20+R), 256 ($10.50 Super-Knockout Heads-Up Shootout), 187 ($3.30 Courchevel H/L) and 208 ($2.20).

    My best performance in a really big (3000+) field was 3rd of 4612 in the Hot $11, but I've only played maybe 500 of those in total, so you wouldn't expect me to have won one yet.
    Last edited by Xopods; March 23rd, 2015 at 04:50 AM.
    Quote Originally Posted by pikachar
    Honestly didn t read OP I just got to say xopods is right on the money
    I design board & card games!: http://www.benefactum.ca

  6. #6
    Staff Sergeant
    Join Date
    Oct 2012
    Location
    Eugene, Oregon
    Posts
    198
    PFO Points
    2020

    Default

    As a mathematician I should love ROI calculations but I can't. The variance in poker is such a strong factor that ROI is virtually useless. Here is how I deal with variance.

    1) I work my way up. Big MTTs seem the place to make big money but there is good money to be made in smaller tournaments. Just watch "Jonathan Little's Secrets" videos to see what is possible. I would want to be crushing the 180 man Sit-n-Gos before I ever got worried about making money in the big MTTs.

    2) I like to play games where I have a high ROI. Then if you have a down swing it is just less profit and not an actual loss.

    3) I never start out assuming it is variance. If you start out looking at your game and game selection you have a better chance of finding a problem that you can actually do something about.

    And just a comment about variance. Number of games and "knowing" your ROI have no meaning without discussing the tolerance interval of your estimate. The tighter the tolerance the larger the required sample size. I like Phil Shaw's discussion in "Secrets of Sit'n'gos" in the section titled "Bankroll, ROI and Variance".

    Good Luck at the tables.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •