Quote:
Originally Posted by FreakDaddy
And how would you do that though? You can only track them for that session. That's why we would do it random, because you'd see them only for that particular session. I can't see why you'd want to give them a particular name.
It would be tough for your program to do, but it could be done. Way to much UI to design and make sense to the user, though, so probably not worth it.
I can see instances where trying to track across sessions is viable, but they are still shots in the dark. For example, I see a few players that pop up every night that play the same style. Something crazy like 50-80 VPIP, 40-50PFR, insane 3bet, and very aggro post flop. After a few hands, I have these guys identified and I can watch them and note that they are taking the same lines as I have seen before. So I make an assumption that it is the same player(s) and adjust accordingly. Still, it is a leap that these are the same guys, and you really couldn't do it on the more "normal" players that have reggish or passive fish stats; just too many of them that play the same way. Your program would have to implement some kind of interface that lets the user tell it which players to combine/rename. That would be way too cumbersome imo.
A better solution for the user is to wait until the random naming thing is implemented, take some notes, and use PT4 or HM2 to alias these players together. So, if you feel confident that they are the same player, make an alias that says random1fromMonday = random333fromWed = randomw888fromFriday, etc. Maybe from their style or notes you have on them. Maybe they always say the same thing in chat. Who knows? That would on the user to decide the alias.
Even without trying to guess the exact player, I am really looking forward to the random names thing. Using that, we can alias similar players together to get a better population estimate based on stats. For example, I can look through old hands and see that these anon players are bluffing the river with X or better Y% of the time. But that includes everyone from the 3/3 nit to the 90/80 aggrotard, so that population estimate is a bit skewed. If I can alias all the nits together, all the passives together, etc, then I have a read that says "this kind of player is betting X here Y% of the time."