TL;DR After writing the whole thing, I have come to this conclusion. Shady Vendor plays for 7-14 points (not considering Pulling the Strings), with extreme flexibility between defense, offense and hand-fixing. I think that the card plays more like a weak 6 provision card than a strong 5 provision card.
But, I admit, I was wrong. I thought that the card would be a solid 6 provision in strength, playing generally around 10-11 points. It can do that, but it doesn't always do that. Instead, if it is nerfed to 6 provisions, a change I still would not be against, I would strongly suggest boosting the base power by 1. Then it plays for 8-15, which is completely reasonable for 6 provisions.
Now, Pulling the Strings, on the other hand, I was completely wrong about. I voted to buff the card to 5 provisions. Having charted out what this card can reasonably be expected to do, yikes, that was a mistake. Pulling the Strings far outpaces the other options of Shady Vendor in a deck built to achieve its condition, and I think even in more normal decks it outcompetes the other options. It is too strong at 5 provisions. Also, Purge is a complete meme, very funny.
Introduction
A post was made today discussing the possibility of nerfing Shady Vendor to 6 provisions. In that post, I was told repeatedly that Vendor was not worth 6 provisions. I think you are all wrong, and that the card really is worth 6 provisions. So, let's count it out. What points does Shady Vendor play for? I will simply go through each option (on the ranged row), and we can find out together how many points the card generates. As I write this introduction, I don't yet know the results, and so we will find out together. I will not be taking into consideration things like vice, or intimidate triggers, or KOB. Not because I think it is unreasonable to expect something to trigger, but because those points belong to those cards themselves.
The first question, how much is a coin worth? Everyone agrees it is more than 1, but it depends greatly on the situation how much exactly. Let's say, conservatively, that a coin is worth around 1.2 points.
Then, how much are we paying? Again, only considering the ranged row, we could be paying 4 or 5 coins, depending on if we are playing off the books. The difference comes out to 1.2 points. So, if we pay for the ability, we are losing between 4.8 and 6 points.
So, in normal cases we are playing for 5.2 – 6 = -0.8 points plus the effect,
and with OTB we are playing for 5.2 – 5 = 0.2 points plus the effect.
Finally, if we are playing Lined Pockets, we are also gaining a coin at the end, so that's 1.2 points we get back. This will result in largely the same math as we would get by playing OTB, however, I think that the coin you get more on top is worth less than the coin you pay upfront. Both because OTB makes the payoff easier, and has less of a chance of overprofiting. To compensate, I will reduce the final score of LP by 0.2.
So, with LP we are playing for 5.2 – 6 + 1 – 0.2 = 0 plus the effect.
All of these variations will be included in the following way:
Normal points generated: -0.8 + effect
OTB points generated: 0.2 + effect
LP points generated: 0 + effect
Bricked
Starting out, what does the card just play for in a vacuum? It has a 4-point body, and profits 1.
Points when bricked: 5.2
That's not very good, but it's not the worst points ever. Nauzica plays for 4 if it is bricked.
Then, the crimes in alphabetical order.
Eavesdrop
If we pay for Eavesdrop, we gain 5 coins, and a bit of deck consistency. I suspect the deck consistency aspect will have great variance in the impact, depending on the ratio of good cards to bad cards in hand and in deck. I will attribute it the power of 2 points. If you pick Eavesdrop through Vendor, your hand is likely not doing super hot, which is why the value I attribute to the effect might seem higher than it might be in a vacuum. If you need your hand unstuck, you pick Eavesdrop, in which case it is likely to give you something than you had. The difference in power between those cards can be attributed to Eavesdrop, and thus Shady Vendor. I suspect that the 2 points is lower than what this value actually is, but I don't think it's much lower on average.
In total, Eavesdrop generates 6 + 2 = 8 points.
Normal points generated: 7.2 points
OTB points generated: 8.2 points
LP points generated: 8 points
Playing Shady Vendor for Eavesdrop results in a net points of 7.2-8.2. That is much more characteristic of a 5 provision card than a 6 provision card.
Payday
If we pay for Payday, we deal 1-5 damage, and generate 1-4 coins. This means the card plays for 5, 5.2, 5.4, 5.6, or 5.8 as pointslam. Let's take the average of 5.4 as something reasonable.
Normal points generated (pointslam): 4.6
OTB points generated (pointslam): 5.6
LP points generated (pointslam): 5.4
Now, you are likely killing something with payday, not just playing it for pointslam. This means the value that should be attributed to Payday is equivalent to the points that would have been generated by whatever you killed. The full range is too broad to quantify, and subject to too much change, but 5 damage is unlikely to kill something really important (unless it's Philippe), so let's say you sometimes kill a 1-point engine, and sometimes a 2-point engine, on average a 1.5 point engine. In this sense, Vendor can be reasonably expected to generate the equivalent amount of points as a 1.5 point engine, after the pointslam effect.
Using the method I used to the value of the Imperial Marine, assuming that the engine that was killed was expected to last another 4-6 turns (that is the amount I suggested in my previous post, minus one), that comes out to an additional reasonable 6, 7.5, or 9 points. Let's just take the average of 7.5.
Normal points generated (pointslam + control) : 12.1
OTB points generated (pointslam + control) : 13.1
LP points generated (pointslam + control): 12.9
Playing Shady Vendor for Payday results in a very poor performance for Vendor as a pointslam card, and an excellent performance as a control card. Played for control, Vendor plays like a weaker form of other 6 provision damage control cards, like gigascorpion decoction, with a body attached.
Pulling the strings
If we pay for pulling the strings, we get something very similar to Payday, except far more complex. Let's start with this.
This is the base data-set I will be manipulating and prodding to find something worth considering. Let's start by noting how the points change as you run more SY gangs in deck and on the board. The more gang tags can be reasonably expected to be run by a deck that plays Shady Vendor, the further to the bottom and right we move. The only thing I am going to structurally change about this data is that I am going to ignore situations where we are only playing 0, 1 or 2 gang tags. It won't happen and is therefore not relevant. The average points for each amount of gangs (in starting deck and on the board) should be the points along the diagonal of each amount of gang tags. So, 6 gang tags can be [0 seize, 6 profit] or [6 seize, 0 profit], and everything in between. In the chart below each gang tag amount has its own color and average. 2 tags is green, 3 is yellow, 4 is red, 5 is grey, and 6 is blue.
Much, much more math could be done just concerning how to weigh these averages exactly. For example, I think you are much more likely to play Pulling the Strings when you can seize a 6-power unit, then when you can only profit for 6. Thus, all the averages are more or less higher than they appear. Furthermore, You are also more likely to play Pulling the Strings when you have 5 tags on board than 2, so the averages themselves should also not be equally considered. However, I do not have access to objective data to determine how these factors should be weighed. Because of this, I will not give a calculated average points per Pulling the Strings in general. If you want to make the calculation, count the body + payment of Shady Vendor [-0.8, 0.2, 0] up with one of the effects of playing pulling the strings depicted on the chart.
If we did assume that all these squares are equally likely (which I don't), Shady Vendor can be expected to play for the following amount of points.
Normal points generated (pointslam): 5.8
OTB points generated (pointslam): 6.8
LP points generated (pointslam): 6.6
This is not super great, but now let's assume you are playing Shady Vendor into Pulling the Strings when it is actually good, so when there are 4-5-6 tags on board
Normal points generated (pointslam): 7,2
OTB points generated (pointslam): 8,2
LP points generated (pointslam): 8
If you thought that this seemed low, you are correct, this was just the pointslam aspect of the card.
Unlike Payday, which killed the engine, Pulling the Strings also gives you the engine, which can allow you to trigger the points gained in addition to the points lost through control (so long as you can activate the condition for the engine). This gives us the following results, assuming again an uninterrupted reasonable 4-6 turns of engine triggers:
kill and gain a 1-point engine: 10 points
kill and gain a 2-point engine: 20 points
Average points generated: 15 points
Important to note is that seizing engines is much more likely on the 5-6 seize squares than on 3-4 seize squares on the chart, and nearly impossible for the 1-2 seize squares. For that reason I am not going to add any engine points to 1-2 seize squares, add only the 1-point engine seize points to the 3-4 seize squares (as one point engines tend to be around 4 power), and add the average points generated to the 5-6 seize squares.
Again, if you want to do the math to calculate the actual average points Pulling the string plays for in general, be my guest, but I am still not sure how to weigh these averages exactly. Let's just say that Shady Vendor into Pulling the Strings, targeting an engine, is very, very, very good. It plays far above 5 provisions.
Now then, with all the caveats and little nuances in our mind, let us assume for a moment that we are equally likely to play Pulling the strings for pointslam as for engine seizing, that is, counting up each square in both the refined pointslam chart and the Engine seizing charts together and dividing by 2, we get this.
Modifying for the 3 possible kinds of Shady Vendor we are playing [-0.8, 0.2, 0], we can see that Shady Vendor playing for Pulling the strings generates far below average points if it plays with 2-3 tags on board, it plays for low amounts of points with 4 tags (unless it captures an engine), and tends to play for a decent to crazy amount of points with 5-6 tags on board (unlikely to happen outside of Collusion decks). This is the play-pattern of a 6 provision card
I also just want to note that I voted for Pulling the Strings to go down to 5 provisions, to make it good again in Collusion gangs. Looking at these tables, I think someone could very well criticize me for that choice. Pulling the Strings isn't just a strong 5 provision card, it's fucking nuts. Just look at the blue squares. It's a payoff card for an archetype, so it deserves to be strong, but this might be too much. I still want to hold out hope that the card is weaker in practice than it seems on paper, but I don't fault you if you think this was a mistake.
Purge
Okay, from an option that was absolutely complex and cracked to something easy and bad.
Let's start by assuming Purge plays for pure pointslam. It deals three damage (+ any witch hunter you have), and kills nothing. The effect is worth three points.
Normal points generated (pointslam): 2.2 (+ witch hunters)
OTB points generated (pointslam): 3.2 (+ witch hunters)
LP points generated (pointslam): 3 (+ witch hunters)
That's just very bad.
Now let's consider you are able to kill something with Purge, and kill whatever the bounty attaches to. The value generated has a truly massive range, and I don't think it can be predicted to what extent each value is likely to become the bounty target. However, I think that you would only play purge if you can deal the damage, get the bounty, and then get around 7 coins for your troubles. Maybe you can do it with Ixora, or Bare Knuckle Brawler, or whatever. 7 coins is worth 8.4 points. In the unlikely event that you do pick Purge, and it works out, you get these results.
Normal points generated (fairy tale land): 10.2 (+ witch hunters)
OTB points generated (fairy tale land): 13.2 (+ witch hunters)
LP points generated (fairy tale land): 13 (+ witch hunters)
That might make it seem like Purge is actually decent, or even good, but this doesn't happen often enough to be counted on. Usually, it can only be expected to play for its pure pointslam value. In which case, Shady Vendor plays for far below its provision cost.
Stolen Mutagens
Last one, what if Shady Vendor plays for Stolen Mutagens?
We have three options, we can choose 'deal 4 damage', 'poison an ally to gain 5 coins', and 'give a unit veil and boost it by 4'. We can ignore the poison on the second option. Insofar as the poisoning generates points, we should attribute those points to the other cards that make that happen. To make things easy, let's assume that 'deal 4 damage' and 'boost by 4' are essentially the same. Either can pointslam for 4, or kill/protect an engine of some sort. That leaves us with two options. damage/boost for 4, or generate 6 points in coins.
Then the question is if we have set-up or not. Let's start by calculating what the points look like without set-up. If we have to choose between 4 points, and 6 points, we should generally go for the
6 point option, unless the 4 point option can kill/protect something else that will generate points. With that in mind, I think it's only fair to attribute those points to the damage/boost by 4 option. That is, if we have to choose between damage/boost by 4, or generate 6 in coins, we can always assume that a choice for damage/boost by 4 comes paired with some sort of engine value. Because it is so uncertain what that value is precisely, let's say it's a 1-point engine only, as 1-point engines do tend to be around 4 power. So, without set-up, the effect can be 6 points in pointslam, or 4 points in pointslam + a 1 point engine in value. I am assuming once more that there is a reasonable 4-6 turns in which the engines does its thing.
Normal points generated (coins): 5.2
OTB points generated (coins): 6.2
LP points generated (coins): 6
Normal points generated (damage/boost): 9.2
OTB points generated (damage/boost): 10.2
LP points generated (damage/boost): 10
Without set-up, Shady Vendor into Stolen Mutagens plays very much like a 5 provision card. A low points floor with a pretty high points ceiling, dipping into 6 provision territory on a stringent condition.
Now, what about with set-up. Because we are no longer forced to choose between damage/boost by 4 and gain 6 in coins, we can also no longer rely on the natural assumption that you would only pick the 4 if it came with something else. So it can no longer be naturally assumed to be an option. So, I will split the card into three sets. Either you can meaningfully damage/boost 0 engines, 1 engine, or 2 engines. that last one is much more unlikely to happen, though. Again, because of the uncertainty, let's be conservative and count these as 1-point engines. So, the effect is either worth 10 pointslam, 10 pointslam + a 1-point engine, or 8 pointslam + a 2-point engine.
Normal points generated (pointslam): 9.2
OTB points generated (pointslam): 10.2
LP points generated (pointslam): 10
Normal points generated (1 engine): 14.2
OTB points generated (1 engine): 15.2
LP points generated (1 engine): 15
Normal points generated (2 engines): 18.2
OTB points generated (2 engines): 19.2
LP points generated (2 engines): 19
As should be expected, going Shady Vendor into a set-up Stolen Mutagens is a beast of a play. If you can kill one engine, the card plays like a very strong 6 provision card, and even baseline it plays like a Nauzica Sergeant used to, which was completely fine as a 6 provision card.
So, what provision cost is this all worth?
First, note that we have only discussed one side of Shady Vendor. The whole melee row is out there, with all the wacky things that can happen there. Just consider this. These effects (except for Purge and Pulling the Strings), are all fair 5 provision effects with some upsides. Do you have a bricked hand? Eavesdrop helps out, do you need to hit something for 3-5 damage? Purge, Stolen Mutagens and Payday are there for you. If you have set-up, which is not too hard if you are playing devotion, then Stolen Mutagens is absolutely cracked as a baseline option. If you are playing KOB, and you need some tribute cards in your deck, then Vendor is the giga-nuts. A super midrangy card that can jack up your coins in a pinch, and acts like a multitool of fair cards.
I am sorry, I think this card is very comparable to Slave Driver. Each individual option is fair enough, even Stolen Mutagens, but the fact that you can essentially run all of them in one card is crazy. You don't need to run eavesdrop in case you have a bricked hand, just run Vendor instead, et cetera. All of this is to say, I think that the ability to flexibly choose (with some rng), all these effects, playing generally in the range of 7-14 points, is worth about 6 provisions. If you disagree, that's fine. In that case, what if we also boost the base power by 1? Then it plays for 8-15, isn't that totally reasonable for a 6 provision utility card? Slave driver plays for around that, and everyone agrees that change is good.
The elephant(s) in the room
You might have noticed that I didn't mention Pulling the Strings there. I think that, even without a card like Pulling the Strings, Vendor is capable of going to 6 provisions without too much damage. However, Pulling the Strings should not remain at 5 provisions much longer. Maybe, just maybe, it turns out that the collusion deck isn't very strong. If that happens to be the case, that will be in spite of the fact that pulling the strings is insanely powerful for its cost. The value fluctuates a lot depending on what is seized, but the sheer amount of points it generates when it seizes something good should make it a 6 provision card.
I also didn't mention Open Sesame at all here, because I wanted to talk about Vendor in its current state. I think that if you nerf Vendor to 6, you could buff Open Sesame to 5 again. The math works out okay. Open Sesame pointslams for 7.4 points, and gives 4.8 points at a later point.
Normal points generated (pointslam): 11.6
OTB points generated (pointslam): 12.6
LP points generated (pointslam): 12
This is similar to a good payday, or a set-up Stolen Mutagens. Which is also why I am not a super big fan of this idea anymore. Open Sesame generally outcompetes the other options at a baseline, which takes away so much from the fun Shady Vendor. You can buff Sesame, but I am not voting for it. And I would strongly suggest against buffing it back without changing Vendor. The pointslam is too strong for the provision cost of Vendor, plus stacking the carryover like this should cost more.
Source: https://www.reddit.com/r/gwent/comments/17lomm9/shady_vendor_ranged_row_math/
