Map vendor formula: should add "2 unidentified maps of same base = +1 level map"

"

Snorkle_uk wrote:

you said no one would craft for profit without eternals, and people do craft for profit without eternals.

I never said this. I did say no one would craft for profit without Mirrors. And I stand by what I actually said.

I will admit I rarely play Standard and was trusting a website for exchange ratios on Alch:Chaos. Actually, two sites. I do find it suspicious that two different sites say 2:1 when you say 4:1, but meh.

But most relevant to the topic, I am not saying high mappers would never use the formula. I have qualified my statements to avoid saying this. That would be like saying "top players" never used Eternals to save a 5L before rolling for a 6L - it was not the primary use, the primary uses where Exalts and Mirror-service Chroming - but some people did it anyway. Just like how some people yolo Exalt. Suboptimal happens.

Nor am I saying a 2:1 vendor formula would be forever suboptimal for high maps. I am predicting it would be suboptimal for 78+ maps, now, in the current mapping economy. This would not necessarily last forever, though. I also am admitting, freely, that it would be OPopie to vendor up your 74-76 maps in the current economy at 2:1 rates.

My point, which you missed, is this: in the instances where a 2:1 map vendor formulas is optimal (and thus OP), there would already be fucked up shit going down. I'm more concerned about preventing that type of shit being inflicted on players, than preventing rampant overuse of a vendor formula.

Just a question: Why is something that is optimal necessarily OP? I don't quite see that.

"

Char1983 wrote:

Just a question: Why is something that is optimal necessarily OP? I don't quite see that.

The definition of "overpowered" in a build-defined game is "optimal in all instances." Because the evidence of power is whether players (always) choose it, which in term correlates with how powerful it is.

"

ScrotieMcB wrote:

Polaris wants math. Okay.

Maplevel, poe.trade lowest price (Warbands, in Alts), ratio with previous maplevel (:1)
68, 1, n/a
69, 2, 2
70, 2.5, 1.25
71, 2.5, 1
72, 2.5, 1
73, 5, 2
74, 8, 1.6
75, 20, 2.5
76, 50, 2.5
77, 120, 2.4
78, 170, 1.42
79, 340, 2
80, 610, 1.8
81, 680, 1.11
82, 1190, 1.75

Average ratio from 72 to 82 is (1190/2.5)^0.1, which is 1.85.

So maptraders currently never get a ratio less than 2.5:1 and very often enjoy better than 2:1 ratio... on average, maptraders currently enjoy a 1.85:1 ratio on upgrading their maps. A 2:1 vendor formula would bring some changes (especially to 75-77 sustainment, the value of 72s would probably about double while 71s would maintain their price) but have virtually no effect on the trade of 78 and higher maps.

Thanks for providing some concrete support!

While trade value does serve as a proxy for sustainability, there are some factors which influence the economy that are not directly caused by sustainability. Since the data you provided is not direct information about sustainability, but sustainability is what's actually at stake, I am going to point out these additional factors.

1) Some levels are more popular than others. For instance, 74 maps are particularly unpopular, which lowers their value. Comparing 75s as a ratio of 74s, the anomalous value from the 74s trickles into the ratio score of the next level, inflating the relative trade value above its relative rarity.

2) Next, it seems safe to presume the total demand for maps is not significantly decreased just because certain levels are worse than others. For instance, the purchasers that aren't buying 74s (causing the low trade value) are still buying elsewhere in the hierarchy. This warps the relative trade values in a way that has no correlation to relative rarity.

3) Lastly and most importantly, these ratios aren't in a vacuum. The 3:1 ratio is already baked in. With trade values as our data points, we aren't talking about how a new 2:1 formula would change things, we are looking at the marginal change from a 50% boost to the efficiency of the 3:1 formula. It sounds like semantics, but what it means is there are more "free" maps for chisels (decreasing their value) and however much these vendors are already in use would produce 50% more maps for the same cost. This will change trade values before accounting for any new behaviors caused by the formula, just because the vendor results from what's already occurring would be improved. How much vendors are already in use becomes very important if we are going to base our decision on trade values. Now I will admit that in higher maps they probably aren't being used much, if at all, but low and mid maps could already have a substantial effect once running through the residuals of the efficient formula. Once new behavior is factored in, the values would certainly change much further.

That was a pretty abstract description so I'll include an example. Let's suppose there are currently 100 L74 maps produced per day by the match 3 vendor recipe, just to put a number on it.
With 2:1 ratio, this jumps up to 150 L74 maps before accounting for any change in behavior caused by the efficient formula. This is "worth" 75*L75, or 37.5*L76, or 19*L77... 150*(1/2^n)*(L74+n)
Compare to the 3:1 formula which was only "worth" 33*L75, or 11*L76... 100*(1/3^n)*(L74+n)
The old formula is not only less maps, it's less maps even trickling up 1 fewer times. This benefit is before even considering changes in behavior. To accurately use the trade ratios, you must be able to control for this effect- adjusting trade values to what they "would" be with no change in behavior under different vendor recipes. The trade value of maps would decrease proportionate to the following factors:
-Map level (higher levels decrease more)
-How much vendors are already in use
-How much more efficient the exchange formula is made (actually not a proportional change, but an exponential one!)

This is a troubling correlation matrix. A misjudged value can have strikingly large consequences, especially at the higher levels.

"

PolarisOrbit wrote:

Thanks for providing some concrete support!

:)

"

PolarisOrbit wrote:

1) Some levels are more popular than others. For instance, 74 maps are particularly unpopular, which lowers their value. Comparing 75s as a ratio of 74s, the anomalous value from the 74s trickles into the ratio score of the next level, inflating the relative trade value above its relative rarity.

Then, compare those 75s to 76s...

"

PolarisOrbit wrote:

2) Next, it seems safe to presume the total demand for maps is not significantly decreased just because certain levels are worse than others. For instance, the purchasers that aren't buying 74s (causing the low trade value) are still buying elsewhere in the hierarchy. This warps the relative trade values in a way that has no correlation to relative rarity.

I don't think you mean by that last sentence that relative rarity ceases to have correlation with trade value, but that there is an additional factor, supply as well as demand. To which I say: should be. But even among maps of same maplevels it is actually spooky how little variance there is in pricing between the good maps and the stinkers. (Notable exception: Malformation is noticeably [edit: about 10%] lower, which I shot right past when collecting my data.)

"

PolarisOrbit wrote:

3) Lastly and most importantly, these ratios aren't in a vacuum. The 3:1 ratio is already baked in.

Spoiler

With trade values as our data points, we aren't talking about how a new 2:1 formula would change things, we are looking at the marginal change from a 50% boost to the efficiency of the 3:1 formula. It sounds like semantics, but what it means is there are more "free" maps for chisels (decreasing their value) and however much these vendors are already in use would produce 50% more maps for the same cost. This will change trade values before accounting for any new behaviors caused by the formula, just because the vendor results from what's already occurring would be improved. How much vendors are already in use becomes very important if we are going to base our decision on trade values. Now I will admit that in higher maps they probably aren't being used much, if at all, but low and mid maps could already have a substantial effect once running through the residuals of the efficient formula. Once new behavior is factored in, the values would certainly change much further.

That was a pretty abstract description so I'll include an example. Let's suppose there are currently 100 L74 maps produced per day by the match 3 vendor recipe, just to put a number on it.
With 2:1 ratio, this jumps up to 150 L74 maps before accounting for any change in behavior caused by the efficient formula. This is "worth" 75*L75, or 37.5*L76, or 19*L77... 150*(1/2^n)*(L74+n)
Compare to the 3:1 formula which was only "worth" 33*L75, or 11*L76... 100*(1/3^n)*(L74+n)
The old formula is not only less maps, it's less maps even trickling up 1 fewer times. This benefit is before even considering changes in behavior. To accurately use the trade ratios, you must be able to control for this effect- adjusting trade values to what they "would" be with no change in behavior under different vendor recipes. The trade value of maps would decrease proportionate to the following factors:
-Map level (higher levels decrease more)
-How much vendors are already in use
-How much more efficient the exchange formula is made (actually not a proportional change, but an exponential one!)

This is a troubling correlation matrix. A misjudged value can have strikingly large consequences, especially at the higher levels.

First off, kudos for starting at 75 and ending at 77. Some reading this thread might confuse this with a 15-long series, but I get you.

Second... yeah, this is the good point. I still stand by my earlier point, that (>2):1 trade ratios only happen when something about maps is already messed up, such as our current debacle with mid-tier maps. However, forcing a new, artificial ratio on an economy is a messy business, and it can be virtually impossible to foresee all possible outcomes.

The best I can say here is: GGG should smooth out midtier maps to return a natural progression. And implement the 2:1 formula, to create an emergency bandaid for future situations similar to our current issues (and also help selffound).

"

PolarisOrbit wrote:

Spoiler

And one last comment that I meant to work in, but didn't find a place for it: a vendor ratio of X:1 maps means the minimum rate of upgrade on any level of map can't go significantly below 1/X. In other words, an L82 map drop can't be too far below 50% from an L81 map. That constrains design in a way that may not be intended, and certainly subverts the notion of investing big currency in your maps when you can get a comparable upgrade by just talking to the vendor instead.

Given Bex's post in this thread I doubt that is developer intent...

It seems reasonably clear, while sad, that GGG don't feel they need a band-aid/emergency fix as the elasticity of trade (demand/supply) translates to an allowance for them to tweak endlessly. Again, in their minds only...

"

davidnn5 wrote:

It seems reasonably clear, while sad, that GGG don't feel they need a band-aid/emergency fix as the elasticity of trade (demand/supply) translates to an allowance for them to tweak endlessly. Again, in their minds only...

It would seem to me that if "Path of Exile's economy is the most important element of the game" to GGG, they wouldn't be so cavalier about pushing it to its limits.

"

ScrotieMcB wrote:

"

davidnn5 wrote:

It seems reasonably clear, while sad, that GGG don't feel they need a band-aid/emergency fix as the elasticity of trade (demand/supply) translates to an allowance for them to tweak endlessly. Again, in their minds only...

It would seem to me that if "Path of Exile's economy is the most important element of the game" to GGG, they wouldn't be so cavalier about pushing it to its limits.

Just as with (im)balance, they can be as cavalier as they want.

Nothing lasts longer than 3-4 months in leagues they care about.

Instead of making ridiculous 20% (what?) boss map drops, let the only bosses drop some kind of map shards. You want people kill bosses, fine, reward them. You kill boss, collect shards, upgrade map, simple. Ratio is up to you. Just stop that fucking RNG content gating.

"

ScrotieMcB wrote:

My point, which you missed, is this: in the instances where a 2:1 map vendor formulas is optimal (and thus OP), there would already be fucked up shit going down. I'm more concerned about preventing that type of shit being inflicted on players, than preventing rampant overuse of a vendor formula.

Fair play, I feel like GGG will be concerned about it though, you know what theyre like. As anuhart once put it, we need to have a shitty map system so we can have a shitty map system right? On some level if the map system wasnt shitty it might as well be a series of static zones, god forbid.


And dont get me wrong, youre a smart guy scrot, Im a smart guy and youre probably a smarter guy than me. I do have a lot of experience tho, my guts been places.


I do feel that 2:1 would never be optimal, for sure it would take something seriously messed up for that to take place. I mean the minute thats optimal maps stop being sold and that in itself is going to mean someone who wants to sell maps has to cut price to make it happen, its self correcting essentially.

I feel like where concerns come in is the new ilvl83 affixes, lvl100 ladder race etc, theres other concerns they have tied into the map system at the highest levels where its not just a question of maps alone. Its like a web of game balance, you start pulling a string here or there and suddenly things are bunching up and getting all freaky somewhere else entirely. Thats where I can see GGG looking at this and thinking "yes, but no, although maybe if..."

But who knows, their intentions and designs are often a mystery.