Observations on 180 Man Turbo MTTSNG Variance: Comparing 2013 with 2014 Leaderboards

My observations given I used the calculator properly, is that in any given year almost all players will not have enough empirical data to know their own winrate.

We might be able to make some useful observations even if we don’t have all the information we need. The OP from the forum post seems to be lacking the pictures.  For now we will concentrate on the numbers from the spreadsheet.  There are 2 years: 2013 on the left half and 2014 on the right half.  Each side is again split into \$2-\$5 stakes on the left and \$5-\$15 stakes on the right.  Once more each of the separated stakes is divided into top 20 profits on the left and top 20 counts on the right.

For each \$2-\$5 section there is the players name, profit, count, roi%, and avg stake.  For each \$5 -\$15 there is the player’s name, profit or count, and then another separate count, profit and roi for each of the \$8 stakes and the \$15 stakes in order that we may see what a player’s separate 8’s or 15’s stats are.

Along the bottom are some sums and averages not much (or at all) referred to.

Observations:

It doesn’t seem to be a big enough sample size to truly compare year to year and decide whether or not the overall profitability of the game has changed.  By the numbers it almost seems the game hasn’t changed much or has gotten slightly better, but it’s doubtful anyone would agree with that (other than Dnegs and PS).  There are many reasons such a comparison might not be accurate-for now we’ll have to make other smaller observations and see if there is any conclusion to be made that way.

2013/2014 \$2-\$5 top 20 profits

We will largely ignore the possibility of \$2 stake games in our data, although, for example in 2013 there is one player with an average stake of 5 that might play a mix of 2’s or possibly not re-buy very frequently, but it shouldn’t bother our analysis to ignore the difference. In 2014 one player has an average stake of \$5 as well.

In 2013 the top two profits for \$2-\$5 stakes were 63k and 56k respectively compared to 48k and 43k for 2014  The 1st place player for both years is “opted-out” and the 2nd place player had 15k games for 2013 and 11k games for 2014 (none are the same players for each year and rarely is this ever the case so we might assume as such unless otherwise stated).  The total profits for the top 20 of 2013 was \$740k for about 37k games  for the average player.  In 2014 the total profit was \$729k for about 36k games on average.

There are 5 players count missing from 2013, the 1st, 3rd, 4th, 14th, and 15. The total games for the 15 players remaining is 177,033 and the average of them is 11,802 games. The 1st, 4th, and 10th, players count are missing from 2014.  The total games for the 17 players is 192,877 for an average of 11,346 games.

The average stake seems to represent some amount of re-buying.  Most players most likely double re-buy initially, always add on, and always double re-buy when if they bust before the add-on. This should amount to a rebuy of 10+.  For 2013 the Avg stake ranges from 5 to 10 with most top regs around 9 or 10.  The average seems to be about 8 for 2013 and 8.8 for 2014.

Roi ranges from the 30’s to 40’s mostly and can go below or above however these numbers aren’t very useful without proper weight.  Some players have 10k-20k games which might be useful for analysis but many have less than 10k which is probably not a very decent sample with 20 players total to choose from.  A 30%+ roi might give a strong enough sample but it’s tough to suggest these are real numbers (players would never load the alternative).  What is likely is the roi is down close to the 8’s and 15’s level but sharkscope’s re-buy tracking does not account properly.

2013/214 \$2-\$5 top 20 count

Less telling for our re-buy’s since most of the \$2-\$5 counts will involve players from the 2’s stakes either mixing or only playing \$2 games.  Fried liver, in 16th place has 15k games in 2013 while holding the 2nd place in profits for the same year. Vasily has an avg. stake of 5 and so likely either mixes 2’s with re-buy’s or doesn’t re-buy with the above stated strategy, yet did in fact make the top 20 profits. madaAK with 15k games made 16th in the profits.  In 2014 Fried liver didn’t make the count leaderboard but did make 5th spot on the profits board.  For the most part players on the profit boards of the \$2-\$5 stakes for each year are not on the count.  The bottom count for each year is 11k and 15k (this suggests not many other players sample sizes are valid since they are not either on the profit board or have a decent sample at a lower roi).

2013/2014 \$5-\$15 top 20 profits

We divided each of the \$5-\$15 stakes (profits or count) again further into 8’s and 15’s in order to get an idea of what the “combined” statistic is made up of. In 2013 the top profit is \$34k by a player that is opted out and \$35k by lijey in 2014.  Lijey played about 13.5 games and slightly over half were 8’s (the rest 15’s).  Their roi for each was 24% and 23%, for 13.5k games it might not be a perfect sample but this player is seemingly strong in the field (perhaps unnoticed as well).

2nd in 2013 to MrNorberto with \$26k profits and 11k games (1st place stats not available).  MrNorberto played almost 16k games, over 10k of which were 8’s with an roi of 16% (19% for 15’s). 10k games is a decent sample with 16% roi but the 15’s sample is lacking and these two players are the top and so can easily be outliers. What is noticeable is the \$10k difference between 1st and 2nd, we would need to have a count for the 1st player but what is likely is they had a great (exceptional) run that year regardless. 2nd place in 2014 is closer to 1st this time with \$32k (vs \$35 for 1st) with about 2k games less (11k).  3rd place in 2014 made nearly \$8k less with almost the same number of games

In 2013 the profits drop off at a fairly normal seeming rate.  What is noticeable is I AmLegend11 with 24k games and \$21k profits (roi of 14%  and 4%) who comes somewhere around 10th spot amongst many players with half the count but nearly the same profits  Boon79s is in the nearly the same boat with a few thousand less in profits and games and further down the line forboon isn’t far behind.   Other than that most players seem to have around 10-12 games. The same non pattern exists in 2014, which many players with ~20k games count  are interleaved with players that played ~10k games.

For 2014 Lijey and Dany871 have the best Roi combinations 24/23 and 20/23 and about 13k games a piece each fairly evenly distributed between 8’s and 15’s.  This might be a minimum sample size although again these are the two top players of the year and they do seem to be outliers.  No-else seems to come near to 20% roi with a decent sample size close to 10k.  It SEEMS like somewhere between 10%-15% with a 10k+ sample size is about the norm and attainable roi%.  However being the top 20 profits in the field it may very well be at least even a few points behind this with an even bigger sample size by at least a few thousands more games than that..

2013 is the same and possibly there are no equivalent scores to the top 2 from 2014 (again 1st for 2013 is hidden).  What is plain to see is there are no players with a strong sample that shows a 20% roi is attainable.  There is little evidence even 15%-20% is a reasonable expectation over a decent sample. Its difficult to tell what amount of players might achieve 10-15%.

For 2013 the only sample for 15’s above 14k games returned a 4% roi, for 11k, 10k, 10k and 11k games respectively for 8’s players had 16%, 14%, 15%, and 6% roi’s. In 2014 no players played 10k games in 15’s and the highest sample was 8100 with an roi of 11%.  For 8’s there was 10k, 10k, 12k, and 10k games samples with 10%, 8%, 12%, and 13% roi’s respectively.

2013/214 \$5-\$15 top 20 count

Count might not be useful in the same way but it is worth observing.  We might wish to look for habitual donators in some games but probably here mostly (and unfortunately) this won’t be the case. We expect to find a lot of winning players and break even players mass tabling.

Count for 2013 ranges from 27k to 13k games with the highest total profit MrNorberto at 26k, I AmLegend11 21k, booone79s 19k, and redhot0007 at 17k (nearly all from 6k 15s). In 2014 the author is still waiting on 15’s stats for 1st place and hotjenny314. Some of the notables are MtZK \$24k, Mikwej \$18k, lijey with \$34k, Dany871 \$32k, N.Ionas \$24k.

The least count for 2013 was 13k and 11k games for 2014.  Both seemed to have the same average amount of games although with missing players its seemingly not a very accurate sample.

Problems

There are many issues with this analysis most of which is the obvious lack of information and large enough sample sizes. Some of the information is not available because players are blocked/opted-out.  The re-buy’s don’t seem very accurate, and the author will look to other players in the field for further explanation.  Most of what can be shown is really just that there is a lack of information.

Conclusion

It’s difficult to make solid over all conclusions with such little information on the overall field, however, there are some important and helpful observations that can be made.  It seems nobody has an roi of 20% or higher with any decent sample size.  The 2 players that came close are anomaly and outliers. Perhaps they are of those players often encompassed with the statement “Anyone that get’s good at 180’s just moves on anyways”, however its difficult to suggest such an roi is attainable or reasonable when only 2 players have come close and not with a reasonable sample.

Some other observation can be made as well.  There are not a lot of player showing up year after year in the top 20 (fried liver).  Also profits are never related to count in an stable manner.  No players serve as solid reference points as some will do great in 8’s and poorly in 15’s or in 3’s or vice versa.  There might be good or better players not on these tables however then they must have had less than the lowest top count or profits and so at best they cannot have a decent sample to suggest such a thing. OOmekatzoOO1 and lijey did show up on both profit boards in 2014 though.

What remains remarkably clear to the author is the overall empirical evidence for 180’s does not at all show whether or not the game is profitable or sustainable and increasing or decreasing over time. What we do hope to show, if it is true, is that clearly one cannot establish an roi with any decent confidence (at least in a years time; even two for most) or at least that there is no evidence to show this.

Players with decent roi’s but with small and bad sample sizes seem to be making the reverse judgement that because of such a high roi, one doesn’t need a high sample.  But without knowing the sample size needed, it is not correct to think that one’s roi is a representative of one’s true roi.  This does seem to be the great mistake, and its not hard to see how players would talk themselves into such self fulling “prophecy”.

Expanded Observations