Article1, percentil & suggestion

[boltrok]

I'm trying to calculate stats over my own database with intervals similiar to the ones showed in Article #1 about leaks.

I don't know how you determine the intervals if you use percentil and/or Quartil.
I'm goin crazy trying to make a excel document that automatically calculates these stats.

< percentil(20)
between percentil (20) and percentil(40)
between percentil (40) and percentil(60)
between percentil (60) and percentil(80)
> percentil(80)

¿¿??

One way to improve could be to add (in Players Tag) a new criteria to the filters. The bb/100 stat to make possible filtering only winning players.

I know you're showing, deviations and this intervals from players tag if you click stat colum but the fact of being imposible to delete losing players make this results almost unuseful.

Thanks for all.

[Rvg72]

Deleting losing players would be a mistake because even a very winning player can have 50K+ hand losing stretches. Using a minimum hands filter said relatively high is much better since the poor players rarely make it past say 5K hands.

[lionzion]

I make my own functions that calculates median bb/100 for some stats.
First you define number of groups and the size of the sample(number of players) for a group to be still relevant. Than the intervals are calculated itself.
After the groups are created, the group with the biggest value(bb/100) is used to make another detailed table with intervals inside this group. This loop make detailed tables for a stata until a group with a relevant sample existed.

If someone is interesed in it I will post my solution here.

[Dronto]

I make my own functions that calculates median bb/100 for some stats.
First you define number of groups and the size of the sample(number of players) for a group to be still relevant. Than the intervals are calculated itself.
After the groups are created, the group with the biggest value(bb/100) is used to make another detailed table with intervals inside this group. This loop make detailed tables for a stata until a group with a relevant sample existed.

If someone is interesed in it I will post my solution here.

lionzion, i'm very interested. Could you pls post it here or even better PM me your spreadsheet. Thank you in advance.