The end of PT, PV, GT+, and other datamining tools

[teddyFBI]

It just occurred to me that if Party and skins ever decide that PT, PV, GT+, etc. are unfair tools, all they'd have to do is allow players to change their playerIDs more frequently than the once every 6 months they currently allow. If players could change their IDs every 2 wks, for example, wouldn't this essentially render useless any and all data mining programs? Any reason party/skins wouldn't do this? all I can think of is that it might exhaust the universe of potential screenames, and people would have to resort to nonsensical strings of numbers and digits as their screenames.

[Lost Wages]

I don't think this is going to be a problem.

1) Party doesn't care.
2) Hardly anyone would ever change their screen name anyway.
3) For PV/GT use, there isn't much added value to having 1000+ hands on a player vs 100. In fact, having only recent stats may actually be more valuable.

On the other hand, they can put a quick end to datamining by simply not storing observed hands on your hard drive. Not quite sure why they started doing it in the first place.

Lost Wages

[excession]

I think they might allow you to change tags more often if stuff like pokeredge takes off.

As for datamining with PT I tend to mostly use it for 30 minutes before I sit down at a table - I don't care if they change their tags daily..

[maurile]

[ QUOTE ]
Not quite sure why they started doing it in the first place.

[/ QUOTE ]
Because their servers were being inundated with hand history requests by email. (Because of PokerTracker.)

[Lost Wages]

That explains the hands where you were seated, but why let you store observed hands?

Lost Wages

[excession]

well it wouldn't be too hard for someone to just write a handgrabber program to grab text from the chat window (see the UB handgrabber we use now for instance) - but I am a bit surprised that they made it so easy and that you can see mucked cards (I guess that is just the default hh info though)

[jasonHoldEm]

[ QUOTE ]
That explains the hands where you were seated, but why let you store observed hands?

[/ QUOTE ]

Because "some people" were running programs grabbing hand #'s out of the chat text on observed tables and then running macros requesting them via email a couple thousand at a time, a couple times a day.

[img]/images/graemlins/blush.gif[/img]
J

[turnipmonster]

[ QUOTE ]

On the other hand, they can put a quick end to datamining by simply not storing observed hands on your hard drive. Not quite sure why they started doing it in the first place.


[/ QUOTE ]

they probably started storing HHs on hard drives because their mail servers were getting pummeled by everyone's autorequests every 2 minutes. I am sure the load on their database/servers dropped tremendously once they started storing HHs (which is a relatively recent addition btw).

--turnipmonster

[Sarge85]

[ QUOTE ]

3) For PV/GT use, there isn't much added value to having 1000+ hands on a player vs 100. In fact, having only recent stats may actually be more valuable.


[/ QUOTE ]

Really?

I would think the more data I had on a player the better. I'm always leary "trusting" data because of the short term luck factor involved.

I'd much rather have a 1000 vs 100 - even that is short term.

Sarge[img]/images/graemlins/diamond.gif[/img]

[Lost Wages]

I think that after 100 hands you could sort players into very-tight/tight/average/loose/very-loose (5 bins). After 1000 hands you could probably sort them into 10 bins. Same for passive/aggressive. My contention is that there is little to no advantage to doing so.

Once in a while you probably mislabel a player with only 100 hands but what's the real harm? Personally, I've never seen a player I thought was loose after 100 hands only to realize after a bunch more hands that he was actually tight.

Plus there is some question as to how accurate older data is. Which is better data to have on a player; 1200 hands collected over the past year or 100 hands collected over the past month?

Lost Wages