So, it's that time again when I could play Sealed or wait for ranked draft. And yet again I've come to the conclusion that I'm better off waiting.

According to FK's great breakdown that should not be the case at 50% win-rate, which I have roughly averaged over the last 3 releases.

So if we focus only on these events and the 50% win-case, he gets to these returns:

Sealed: +295 gems
Bo1 Draft (paid with gems!): +209 gems
Bo1 Draft (paid with gold): -41 gems

He has reached these values through these evaluations:
1 gem = 1 gem
1 regular booster = 200 gems
1 Draft/Sealed booster = 577 gold = 115.4 gems

Now while the first two are obvious, the last one carries a lot of assumptions and cannot be "correct" for everyone. He states as much and I assume that he has found the best possible value.

However I want to avoid determining this value which messes with the entire event evaluation process.

My metric is this: How many gems do I pay for 1 Draft/Sealed booster?

I'll be using Frank's numbers here:

Sealed costs 2k gems and he has calculated an average return (still 50% win-rate) of 1,002 gems and 3 boosters, so 1,602 RoI + 6 Sealed boosters. That means I'm paying 398 / 6 = 66.33 gems per Sealed booster.

Bo1 Draft costs 750 gems and returns 347 gems and 1.33 boosters, that's a return of 613 gems + 3 Draft boosters. Therefore one Draft booster costs me 137 / 3 = 45.67 gems.

Bo1 gold Draft (that noob trap everyone always advises on reddit) costs (137+250) / 3 = 129 gems per Draft booster. Not the end of the world, but one could argue that the wildcard portion of a booster is worth more than 71 gems, which makes this event have a negative RoI.


So not only does one Draft booster cost me less than a Sealed booster, it is also more valuable for the purpose of acquiring rares: It's not too optimistic to expect 4 rares on average per draft, so you get 1.33 rares per Draft booster compared to only 1.00 in a Sealed booster.

When I factor that in, I'm only paying 45.67 / 1.33 = 34.34 gems for the equivalent of a Sealed booster in Bo1 Draft and at that point Sealed is almost twice as expensive.

So I cannot agree with the notion that Sealed has a better RoI. I could only see that when factoring in time as a resource - for a Sealed event should take about as much time as a Draft. Also, it's available now, which is certainly a bonus.



Now that I've written this down, I'd like to get these thoughts to Frank. Maybe he sees a flaw that I don't see. Maybe you guys do

edit:
I have willingly ignored two factors that I assume are insignificant and will balance each other out:
1 Draft booster does not grant 1.33 times the Vault progress of a Sealed booster, so with this in mind it's value should be slightly less than 1.33 Sealed boosters.
However expecting only 4.00 rares (mythics count too!) per Draft is on the low end of expectations. Increasing this would raise the value of a Draft booster beyond 1.33.

And given that a Draft booster is already cheaper than a Sealed booster, this isn't that relevant, it only determines how much cheaper it is.

Don't have time to analyze this, but if you're interested in Frank Karsten's response, tweet at him. I don't know how responsive he is, though.

This is relevant to me as I stockpiled gems but only have enough left to do either 1 sealed or traditional draft. I want to get more exposure to Theros limited before using my last reserves. My anecdotal experience is there's more noobs during the Sealed release for a new set, so it's easier to do well in sealed versus draft where there are a lot of really skilled drafters.

I personally am better at building Draft decks than Sealed so I try Sealed like maybe once or twice then wait for Draft because Sealed will drain my gems otherwise.

4 rares a draft, on average, seems high, unless you are taking every rare you see regardless of need.

That would lower your win percentage, lowering your prize pool. And it has diminishing returns the more you play, especially if you open any prize packs.

Yes, taking every rare is implied. Economically that's the right call. It can weaken your deck of course, but my 50% rate is despite that. I'll get back to this later with my stats when I'm home.

Diminishing returns sure, the more rares you have, the more blanks you get, but that applies to Sealed boosters as well. At a certain point you stop drafting and buy boosters that are duplicate-protected.



I also expect Bo3 Draft to be harder than both Sealed and Bo1 Draft, so I haven't taken that into account. If one could maintain the same 50% win-rate there, it'd be even better.

The "noob trap" is the only way to convert gold to gems.

Yes, but the negative RoI (influenced by win-rate) makes it inefficient. That's not my personal interpretation but in FK's charts as well.

You lose some pennies on the dollar for every gold draft you enter and then you can gain them back by purchasing the mastery pass and reaching the point where it's rewards outweigh it's cost by the amount that you lost before.

If anything, with no wins counted, the "Noob trap" is an efficient way to get 3 packs at a 500gold discount. If you have a win rate of any kind then your getting more for free.

I dont think you can just dismiss the argument against 4 rares per draft with, (summarizing) rare drafting is the point and your 50% win rate is proof enough that rare drafting doesnt hurt your prize pool.

It does. You'd be winning more if you dont rare draft. Winning more means more prizes. More prizes means more free drafts. In the long run, taking 2 rares per draft just might mean more rares in the long run but I'm not going to do that speculative math.

And if you stopped rare drafting, then maybe your win rate would be high enough that this noob trap as you call it becomes even more luctative.

Furthermore, efficiency is not the point. If ranked events are the ONLY way to convert gold to gems then that is, in and of itself, the only point up for discussion. RoI, as you put it, is based on win percentage. So stop rare drafting.

How do you calculate that 500g discount? Are you implying that a rare booster is equal to a shop booster?

And you would have to do that very speculative math to make such a claim. I don't see passing a rare - which means losing X gems on the spot - to be a winning move. You'd have to win X more, at least, to make up for that. Not very probable imo.

1 (8 card) pack costs 1000gold in the shop.

Draft is 2500 for 3 (14 card) packs. Not counting the 6 extra cards that is a 500gold discount.

There are simplifications made, but yes, 3 packs at a 500gold discount.


And I dont need to do math to tell you that rare drafting leads to a lower win percentage. You can call it anecdotal and dismiss this offhand, if you want, but you'll be doing yourself a disservice.

I'm pretty sure that Ryan Spain has all this figured out on a excel sheet.

Well, this isn't simplified but plain wrong. The entry fee is 5k gold for the gold draft. If you account for the guaranteed earnings of 50 gems (250 gold) and 1.2 packs (1.2k gold) you still pay 3,550 gold for those 3 draft boosters.

And the bigger factor is that a draft booster is worth less than a shop booster for you don't get 1/6th of a rare wildcard. Not to mention the default chance of converting a card into a wildcard and duplicate protection.

Now how much value you attribute to that wildcard portion is a controversial topic and this is the aspect I wanted to avoid for you can't put a value on it that's accurate for everyone. One very, very rough estimate is: actual cards 50%, wildcard portion 50%. That of course is super-simplified.

Frank Karsten has put a 577 gold price tag on a sealed/draft booster. That means, he values the wildcard portion at 423 gold (+ a few pennies for vault progress).

So with those numbers, you pay 3,550 for 3 boosters valued at 577, which is 3x577 = 1,731. Ergo you lose 1,819 gold per entry at 0 wins.


This is not the point of contention. I'm not that stupid
Of course rare-drafting has a negative impact on your expected win-rate. I just argue that the increase in win-rate is worth less (in gems) than the card you passed on.

I don't know that guy, but I've googled him, so sure, that's a possibility. However I don't have access to his files, so FKs article - and previous iterations contributed by other people - was all I had.

And that never quite matched my own sheets, so I'm glad that I've discovered the difference and thought I'd share it. Maybe I'll even register on twitter so I can try reaching out to FK.

He might be interested because imo this is an angle he left out. His graphs are based on a one-time entry into a single event - like you'd calculate when you have time for one event and they all take the same time. Then you want to have the absolute biggest gain in gems independent of a relative gain compared to the cost of the event. This approach also requires that 577 gold evaluation.

My findings are that I pay less gems (about half) for such a booster through the draft event vs the sealed event but that it also takes about twice the time because I'm buying 3 draft boosters at a time whereas sealed has 6.

So if time is a bigger concern for you than gems, go for Sealed. If you want to be gem-efficient, use them all on bo1 draft - unless you're a world-class player, then it's bo3 draft.

And if you're f2p, by all means, enter the gold draft if you like limited. I did the same when I was f2p. And I also complained on the WotC forums about the fact that it's a worse investment than buying pure boosters. That still is a shame imo. Buying into an event should always be a better RoI than buying straight boosters because it keeps people engaged in actually playing the game. But that horse is as dead as it gets, so we have to live with gold draft being inefficient economically. Despite what other people keep repeating

He runs a Twitch channel on a FT2 account playing only limited.
Going optimal is the name of the game.

Yep, its 5000g not 2500g.

How much value do you put in gold?

Thank you. I can replace my crude sheets with that

You will get my exact numbers within the "limited event" tab, when you replace the values in column D with the total spoils for each outcome. That is actually very simple because you only need to add the values of boosters. For Sealed just add 600 to each value for 3 boosters and for draft there is a different decimal value for each outcome.

So at 50% win-rate the total cost of a Sealed Event is 997.66 gems per his original values and 397.66 if you factor in the 3 shop boosters you "win" at 200 gems per. It is noteworthy that Sealed kind of forces you to purchase 3 boosters for 600 gems because the whole point of this exercise is to pay less per rare than through these boosters.

For bo1 draft (paid with gems) the corresponding values are 402.73 and 137.24.



5 gold = 1 gem because a booster is 1k gold and 200 gems.
The only exception is bo1 draft and that's why that's so enticing to pay with gems. You get more bang for your buck because 750 gems/5,000 gold is a different ratio than everything else.