Software stats you do NOT want in Tournament Poker.

Question for poker calculator programmers: So why it is not possible to create some player account into Tournament Indicator? This way stats would be split. Or another solution would be if TI was smart enough to recognize the type of game at least the type of table (2 6 9 10 handed table) and this way discriminate the stats.

Turtleknife explains the tournament statistics in poker software - I am reminded now about this, because I thought of suggesting such segmentation before. Then I realized, at least for TI that it was not useful for several reasons.

1- Tournament profiling is all about what is happening at that very moment. Historic data is useful for knowing how players react in certain Ms, but as for the table size or what site they play it, is completely irrelevant for the game, structure, and payout at hand.

2 - You should always be on the current statistics, and NOT the historic stats when using TI. The current stats do not reflect any past data you had on this player, because it is likely not useful or comparative to the situation at hand. The only time I use H is when I am joining a table and entering a hand right away, or I want to have a look at their M data and how they react in different situations.

3 - In tournaments if you divide the stats in the myriad of ways you are suggesting here, you just are not going to have anything useful on the majority of players because you simply won't be sitting with your opponents frequently enough. There will be a few exceptions, but for the most part if you divide by site, by table, by turbo etc, the data will be sliced into uselessness.

Then another question to Marty: OK, I am going to take my life in my hands here - it is only my second post, and I am going to ask Marty to clarify this for me :? I have always used the historic stats. My logic is (probably 'was' after I see the reply) was that any individual game can have huge variance.

For instance, TI says I am TAG and I am sure I am. However, I can appear in some games as a maniac as the first 10 hands I have had great starting hands - if only, but you know what I mean. Conversely, I can look like a mouse as I don't pick up a card higher than a T and all off suit.

Using the historical stats, would this not 'smooth it out' and so you would see the true characteristics of the player, rather than a snapshot? Why is poker so difficult !

Turtleknife responds again here: Theory is one thing, practice, quite another. You are just not going to have enough data on any particular player to make it useful to a given situation. I might check H data when I come to a new table to see if I have played with the dude before, but I always revert back to current stats for several reasons.

The situation you are currently in is unique, in MANY ways. Tournament stage, blinds, payout structure, other opponents at the table, Q, M, position.... shall I go on? You will not be able to extract data applicable to your current situation, no matter how much you have played an opponent before. This is an important distinction because you could very well be comparing stats of a player in a 5,000 deep stack tournament with his stats from a 300 starting chip super turbo!!! Extreme yes, but can you see how the disparity in such tournaments could skew data you are relying upon now?

Are YOU not a different player in different circumstances? If a player is not different under different tournament circumstances, then they are simply not winning.

Now that being said, T&F is right about long term stats for cash games because if a player is a regular at certain tables, many of those factors are smoothed out as long term characteristics under similar circumstances, but this is just blue-sky-dreaming for tournaments though. You are never going to get comparable data. Which is why it is fundamentally important to give your profile weight to what is happening at the table and the tournament as you are playing in that moment.

That doesn't mean to say you don't want to know the player's tournament history though, but there are 2 better ways to do this, I know of. Firstly, within TI you can click directly on a player's name and get his MZone data. If TI has collected stats in different M's on him during this or any other tournament it will divide and show in a pop-up window. This will give you clues, if he can adjust accordingly. - so at least you have scoped one important historic piece of data with relevance to to his M.

The other way is to check his profile at OPR and look into those numbers to see where he performs well, how deep he gets, and if he is profitable or not. The data there to me is much more concrete and you can at least extrapolate to figure if you are playing a monkey or a genius. If I am getting deep into a tournament, I use OPR frequently for searches. Early, I rarely bother, because it just doesn't matter much to me. I know all I need to know with current stats in TI.