Understanding Map Drops - a statistical Approach

"

tackle70 wrote:

"

Johny_Snow wrote:

The strategy is to not stack quantity on your gear because the game is rigged like that.

Quantity on gear does not affect map drops and never has. Map drops are only affected by quantity on the map itself. This has always been the case.

Not quite.

0.9.12c Patch Notes

"

Chris wrote:

Version 0.9.12c
End-Game Map Changes:

• The player's "Increased Item Quantity" stat does not affect the drop rate of Map items any more. It does still increase the drop rate of everything else inside Maps.

Very interesting read. Are you 100% sure about the gaussian distribution?
I make a statistical survey on my map drop, based on the itemlevel of the map (i.e. i did note the itemlevel). At this moment I only have a small sample (176 records), but I'm still logging drops. I started logging in the second half of march 2016.

I follow these rules:
1) I do not note maps dropped outside maps (i.e. non-Zana masters, lab runs, etc...)
2) I do not note maps dropped inside the Zana mission 'find the map'
3) I do not note unique map drops. However, map dropped inside unique maps have been noted.

For each map noted, i logged the level of the drop area (i did not use it in my statistic, but maybe it is useful for furter analysis), the ilv of the map (which can be +0,+1 or +2 with respect to the area) and the tier of the drops (which can be from 1 to ilv-67).

For each map I subdivide the x=0-1 interval in (ilv-67) equal sub-intervals, check which interval corresponds to the tier of the map, and consider 2 functions f(x) and g(x)
f(x) is 0 if x<middle point of the sub-interval, 1 otherwise
g(x) is 0 if x<min x of the sub-interval, 1 if x>max x of sub-interval. It is linear from 0 to 1 in the sub-interval

Basically f(x) is a step function, g(x) is similar to f, but it is continuos. I sum all functions for all maps dropped and then diveded for the #number of samples, getting a total F(x) and total G(x).

Results seems to shows that drop rate is (quite) equidistributed on the tiers (even if it seems that higher tiers have a bit more chance), with the exceptions of the highest and lower(=1) tier. The highest have very high probability and the lowest have very low probability with resect to other tiers. Although the sample is too small to reach a conclusion.

Graph of F and G:


Calculations on excel, with also istograms about tier with repsect to ilv of the map:

https://dl.dropboxusercontent.com/u/23032932/PoEMapstat.xlsm

EDIT: I play on standard

 Ok, so you can do a bunch of statistic math and such and produce pretty graphs and charts. In the end it doesn't mean nothing to the average player (me). I put down my money for a Kiwi pet (CB PoE) and more stash tabs (48 in total) so I have paid GGG over $150.00 USD in supoort of this "free to play" game as I understand that GGG is a company and has to generate revenue or go out of business. I have been in PoE for over 3 years. My first post was on December 26th of 2012 and to date have a total of 2766 posts (hmm.. 2.04 per day). I've survived PoE: The Desync Years, all kinds of realm DCs, countless one-shot bosses that should have been enough to make me rage quit permanently, etc. etc. I've been a PoE loyalist as well as anyone can ask for, and all I ask in return is a "fun-to-play endgame.

 So all the fancy statistics in the world can't satisfy me for this simple fact: I have a level 82 build and this is what I get for map drops:

 I have never gotten a tier 7 or better drop in about 100 maps played this league and only a couple of tier 7 and one tier 8 map for all previous leagues running a level 88 build. That is such an appalling map drop rate that shows many, many, many a frustrated player that GGG doesn't give a damn about making PoE "fun to play" in the endgame. All they want is to gate the higher tier maps behind horrifically low drop chances in an effort to slow down the speed freak PoE elites from level 100 in under 1 month and that didn't even work as at least 2 players got to level 100 in 2 weeks in Perandus League.

 So I play tier 5 or less maps and get bored with face rolling everything and only get a speed bump challenge at the end map boss (if the map even has one). And GGG has the gall to call PoE "working as intended" in the endgame and "fun to play" running maps. I don't know what passes for endgame "fun to play" in the southern hemisphere but constantly being forced to run low tier maps because the higher tiers don't drop is not my idea of fun-to-play. Also, sorry GGG, not getting higher tier map drops is going to get me to trade for them. I don't come here to play PoE: The arpg commodities trading sim game.

 So go ahead and show off your math statistics knowledge and tell us the % drops of all maps for all players over time. Only math stat junkies will ever consider that "fun to play".

"

Arrowneous wrote:

So go ahead and show off your math statistics knowledge and tell us the % drops of all maps for all players over time. Only math stat junkies will ever consider that "fun to play".

Way to shoot yourself in the foot with strawman arguments, pro tip: the OP is in your side.

Probably because you are mathematically illiterate and/or didn't read the OP you thought otherwise.

"

HellGauss wrote:

Very interesting read. Are you 100% sure about the gaussian distribution?

No. It is certainly not flat, though. I've not had time to have a look at your calculations so far, maybe I can do that later.

"

Arrowneous wrote:

So go ahead and show off your math statistics knowledge and tell us the % drops of all maps for all players over time. Only math stat junkies will ever consider that "fun to play".

Map drops are a bad system. I've always said that. They have been so annoying to me that I started this statistics.

I am sorry if you are not able to understand that, which btw is perfectly fine (I am a bit of a math nerd, and not everyone has to be one). The general conclusion I have is relatively simple: Too much RNG.

Sorry for your poor experience. This league (Perandus), map experience for me is pretty good, but that has a lot to do with the fact that I've been able to run and kill Uber three times. And I think it is kind of ridiculous that I had to kill Uber in order to get any useful map drops.

Hmm I think Perandus is a shitty league for map drops. I had no problem sustaining higher maps in last 2 leagues, well, higher in this case meaning t9 - t11. In standard running same maps I get more. Could just be dumb luck tho.

Wherever anyone says "too much RNG," a self-found perspective is at least partially assumed. There is no such thing as too much RNG when one is referring to what the collective might of a trade-connected community can overcome.

So here's a more accurate appraisal: self-found maprollers spend more currency trying to roll map mods, aiming not merely to break even on maps over a million map runs, but instead to never experience a lapse in sustainment (in other words, they're playing around minimums not averages). Then after spending extra trying to avoid the bad luck streak, if they keep playing it will still "inevitably" occur. This effect becomes stronger and stronger as one climbs map tiers, until near the end of yellow maps the game is practically screaming at the player TRADE, MOTHERFUCKER. OR BY GOD THERE'S A 99% CHANCE I WILL SMITE THE SHIT OUT OF YOU.

Is that bad? I don't know. I think there are advantages to basically strong-arming the least casual, most established players into being economic participants. I imagine it'd be a lot more difficult to trade for cheap gear otherwise. I definitely have enough respect for it to be hesitant to change it without a good plan to deal with likely economic ramifications.

So you're saying that GGG wants to make it a necessity to trade for higher level maps or else quit? Strange that GGG is ok with alienating millions of solo players that don't enjoy trading at all and play for the monster bashing, build experimenting, and leveling challenge. Hmm... that must explain why millions have quit over lack of maps. I wonder, does GGG get more from players trading for maps than they lose from players who have quit from lack of higher tier map drops? How does this help their financial bottom line knowing that map starvation is driving away many players?

I don't think it drives that many away. Although it is effective, it's also subtle; most players wouldn't realize how it is working.