Poker Tracker stats for tournaments

[jedi]

How do you guys read your Poker Tracker stats for tournaments? Because the blinds/limits/number of players are constantly changing, I find it hard to make heads or tails out of my stats or other people's stats for that matter.

How do you guys use it?

[Abagadro]

I usually just use it to see how I am doing on Return on Investment and average finish. It's also good for reviewing specific hands. Otherwise I don't find it particularly useful for tourney stats (although my tourney DB isn't nearly as big as my ring, so that may be a factor).

[wayabvpar]

The only real use I have found for tourney stats (other than my own) is to see which of my opponents end up in the money...a few money finishs (finishes?) is a small indication of some sort of competence. Usually. Did I equivocate enough? [img]/images/graemlins/grin.gif[/img]

[Benal]

I have stats for about 3800 SnG players. Of those 3800, about 3750 of them have played only one SnG. The highest number of SnGs I have for a single player is 4! Most of these are at party 10+1

[FloppedFlush]

In addition to what the others have said, I like to keep the game window open, load the new hands in every 20 hands or so, and take a quick glance to see if any numbers really jump out as being extreme. From paying attention at the table, you can tell that a particular player is involved in a lot of hands, but to me, there's a difference in playing "a lot" of hands and playing 85% of the hands. And I wouldn't notice exactly how big the figure was without PokerTracker.

Similarly someone who's not playing much will blend into the background, but it can be interesting to note that someone's only played one out of the last 40 hands and that the one time he played was with KK. You can throw that fact in the back of your mind until you see him start to loosen up and maybe target him for a blind steal if the situation comes up, or know to beware if they suddenly play a hand.

Just don't give a whole lot of weight to anything from past tournaments becasue situations vary greatly. If someone has an extremely low % of hands played, it could be because they were playing tight to get into the money at the end of a satellite, or because they were disconnected while you were at the table. At the beginning of the tournament, I usually use the filter to show only hands up to 50/100 blinds. Then as the tounament progresses, start deleting the earlier blind levels and adding in the later ones so the results are centered around the level you're currently playing. A lot of players will change styles depending on the blind levels, and if you can get a decent read on a player from times you've played against them in the past, then some information is better than nothing. Just don't rely on it too heavily - any read you pick up here is going to be a lot less accurate than a read from a ring game becasue the situations vary so much (limit vs. no-limit, rebuy vs. no-rebuy, SNG vs. MTT, etc.). If you have enough data, you could filter on these variables also, but I don't at this point.

[Bozeman]

This changes quite a bit at higher levels: I have 2000 opps in my data base, and several have played 10-12 games against me. The biggest loser is down 1600 over 12 games, and the biggest winner (besides me) is up 2000 after 5.

That said, I don't use pokertracker much at all.

For personal stats, I think you should look at 1st vs 2nd vs 3rd (I like to also look at 4-10th individually), win rate by level, win rate by position, VP$IP, aggr. factor.

However, because of the relatively small samples you have for other players, and the changing blinds and table size, these sort of average statistics are not particularly good. Therefore, most of my analysis is based on individual observation and not on PT.

Craig