Party Mini Steps Profitability Analysis (In-Depth)

[tshort]

The topic has been debated on here: are the steps worth the time/money? It is seeminlgy obvious that a good player should profit, in the end, off the steps. The question: is the amount of time and tourneys played worth the profit?

To figure this out, I wrote a script (in Python) to run through scenarious of exhausting a large number of entries at certain levels to see how many tournaments one would play and how many entries to Step 5 one will accumulate.

I had to make assumptions about finish probabilities. I do not have near enough data on the Mini Steps to come close to having accurate probabilities of various places at various steps. Therefore, I spent some time reviewing my 11s - 33s data and mocked up the following probabilities of moving from step to step for a great player (you might reference Party Poker's web site to see Mini Steps place breakdown):

Probabilities of moving from a step to another step:

oneOne = .52
oneTwo = .2
twoOne = .29
twoTwo = .38
twoThree = .32
threeOne = .28
threeTwo = .26
threeThree = .25
threeFour = .16
fourOne = .06
fourTwo = .09
fourThree = .20
fourFour = .38
fourFive = .27

Variations in actual probabilities from these mocked values are likely so this will be considered ballpark for great player. I would then rerun the simulation skewing these probabilities down 10% and 20% to simulate a Good player and a Okay player. I assume that expected winnings at Step 5 are $570. This would represent a great player, but my two levels of skewing should account for worse winnings.

Using law of large numbers in my recursive script, I was trying to obtain an overall profit per tourney played when buying in at different levels. I first assumed if you buy in, you exhaust a buy-in until you have lost out or won money. At the end, you would have no remaining freerolls at any steps.

Buying in to Step 1 and exhausting those buyins

How much money would I make if I bought in to ten million Step 1s, exhausting freerolls? How many tournaments would I play in the process? How much would my profit per tourney be?

10,000,000 Step 1s assuming Great Probabilities:
49,270,000 Total Tourneys Played
341,000 Step 5s played
$2.74 profit/each tourney played

Assuming Good Probabilities:
34,726,000 Total Tourneys Played
146,000 Step 5s played
$0.68 profit/each tourney played

Assuming Okay Probabilities:
26,720,000 Total Tourneys Played
63,800 Step 5s Played
$-0.88 loss/each tourney played

Conclusion: Buying in at step 1 is a waste of time. Making $2.74/tourney (at best) isn't worth your time. It will take about 29 buyins at step 1 per step 5 you play.

Buying in at step 2 and exhausting all freerolls

Great Prob: $3.90 profit/each tourney played
Good Prob: $0.78 profit/each tourney played
Okay Prob: $-2.34 loss/each tourney played

Buying in at step 3, exhaust freerolls

Great Prob: $4.20 profit/tourney played
Good Prob: $-0.50 loss/tourney played
Okay Prob: $-6.00 loss/tourney played

Buying in at step 4, exhaust freerolls

Great Prob: $21.00 profit/each tourney played
Good Prob: $16.00 profit/each tourney played
Okay Prob: $7.00 profit/each tourney played

As you can see, it would be logical to buy in only at the 4th step for a decent profit for your time.

What if we buy in at higher steps and disregard the freerolls we receive for the low steps?

Buying in at step 4 and disregarding step 1 freerolls

Great Probability: $32.00 profit/tourney played
Good Probability: $21.00 profit/tourney played
Okay Probability: $8.50 profit/tourney played

Disregarding step 2 freerolls also:

Great Probability: $38 profit/tourney played
Good Probability: $24 profit/tourney played
Okay Probability: $9 profit/tourney played

Should we disregard step 3 freerolls?

Great Probability: $41 profit/tourney played
Good Probability: $24.4 profit/tourney played
Okay Probability: $7.60 profit/tourney played

It would be a close call. If only you could sell them.

Buying in at step 3 and disregarding lower freerolls

Great: $1 profit/tourney played

Lets try only disregarding step 1 freerolls:

Great: $5.1 profit/tourney placed

A little better than playing step 1 freerolls.

Step 2, disregard Step 1 freerolls

Great: $4.50 profit/tourney played

Also, a little more profitable than playing step 1 freerolls.

Conclusion

Steps definitely allow you to have a good ROI, but is that ROI worth your time? It is very possible to have a ROI of nearly 200% by buying in at step 1. That's what lures players to the steps.

I'd bounced around the steps for a few dozen tourneys, then decided to do this analysis to see if they worth the time. It turns out my ballpark analysis would suggest only buying in at step 4 to make a decent profit/tourney. Some lower buyin levels could compare to playing 11s, but could not be as profitable as 22s/33s in terms of profit/tourney.

If only you could sell your freerolls.

If I get a few more hours of free time I will tailor this towards the Original Steps and Steps Higher.

Questions, comments, criticism are welcome.

tshort

[adanthar]

Nice analysis. However, the Step 1's should theoretically have a much bigger return using proper strategy.

Do they still break down (1), (2-6)? Do they still never have over 6 people left at level 4? If both are true, simply fold every non-AA-QQ hand and fold your way into the freeroll, *then* play for first.

You should be repeating or better well over 90%.

[meow_meow]

Can I ask you how you ran this? I'd like to do one of my own.

Also, I think you made a typo in the 3to4 probability - just 0.16 when top 2 advance...

[zipppy]

[ QUOTE ]
Nice analysis. However, the Step 1's should theoretically have a much bigger return using proper strategy.

Do they still break down (1), (2-6)? Do they still never have over 6 people left at level 4? If both are true, simply fold every non-AA-QQ hand and fold your way into the freeroll, *then* play for first.

You should be repeating or better well over 90%.

[/ QUOTE ]

I fooled around with ministeps for awhile (until I won a step 5) and did exactly that. You can basically fold your way into the top 5 in a ministep2 as well (guaranteeing a repeat at the same level).

[tshort]

[ QUOTE ]
Can I ask you how you ran this? I'd like to do one of my own.

Also, I think you made a typo in the 3to4 probability - just 0.16 when top 2 advance...

[/ QUOTE ]

Looks like I typed Step 3s backwards. I will double check my script to make sure I didn't have them backwards in there. That would obviously change results. If you know how to program I can send you the python script. Otherwise, I can see about putting it on the web or making an executable.

[Dallara]

[ QUOTE ]
Nice analysis. However, the Step 1's should theoretically have a much bigger return using proper strategy.

Do they still break down (1), (2-6)? Do they still never have over 6 people left at level 4? If both are true, simply fold every non-AA-QQ hand and fold your way into the freeroll, *then* play for first.

You should be repeating or better well over 90%.

[/ QUOTE ]

Yes, this strategy has worked quite well for me as well (ok, maybe slightly looser, but not much). There are times where you'll get a Mini-2 with 6 or 7 people left at level 4-5, but not too often.

I think the main issue with the mini-steps is the time it takes to move up on this strategy, but since I'm not really optimizing $$/hr, I don't mind it too much. I have started to use my mini-step 1's as an opportunity to employ the Gamboholic strategy though which reduces the time you waste playing them and optimizes for an "only the winner moves up" tourney.