The top three theory

Looking at the bar chart of my stack size over the course of last night's session, three winning hands stand out. On the first one, I gained $19,800; on the second, $10,200; on the third, $9,400. After those three big gains, there's a steep drop-off; my next highest gain is a mere $2,700. Three is a very powerful number, and has inspired me to come up with my latest poker theory; I'm calling this one "the top three theory". It's simple to state, and will be simple to verify (once I get around to it :-). Here it is:

If the top three entries in the sorted list of the absolute values of your hand deltas all come from winning hands, then there's a high probability that the session itself was a winning one.

Here are the top 10 entries in the sorted list from last night's session (note that I've left the negative sign in front of the losses to make it easy to distinguish them from the wins, though the sort was based on the absolute values):

19,800
10,200
9,400
-6,400
-4,400
-4,300
2,700
2,400
-2,400
2,200

During current Hold'em session you were dealt 79 hands and saw flop:
- 12 out of 14 times while in big blind (85%)
- 11 out of 15 times while in small blind (73%)
- 25 out of 50 times in other positions (50%)
- a total of 48 out of 79 (60%)
Pots won at showdown - 6 of 7 (85%)
Pots won without showdown - 7

delta: $20,700
balance: $3,665,639

update: I wrote some code to test this theory, and it held up; in the last 116 sessions I've played, 34 have the top 3 absolute value deltas from winning hands, and all 34 were winning sessions.