No firm methodology at this time. I won't bet until / unless there is one; it might be kinda vibesy.
Buckets are left inclusive, right exclusive, so a price of exactly 160 M/$ resolves to the 160-200 bucket. If there's uncertainty I will default to averaging out and taking the central estimate as best I can.
I will consider things like:
Mana purchase price available in the store, both current and recent sales
Apparent mana purchase patterns (did a sale increase sales?)
Cost of features available for either dollars or mana (for example, if you can buy a subscription with either currency)
Implied exchange rate from charity giveaways
The above list is not intended to be exhaustive.
Some recent reference points:
Black friday sale had M120000/$750 available, for a rate of M160/$. This didn't seem to generate a huge sales spike.
Charity currently lets you buy tickets at 254/M100, with a pool of 2.9e6 sold and $1000 for the giveaway. So the current implied price is M1153/$.
1,000🏅 Top traders
| # | Trader | Total profit |
|---|---|---|
| 1 | Ṁ154 | |
| 2 | Ṁ104 | |
| 3 | Ṁ55 |
People are also trading
Resolution: 160–200 (central estimate ≈ M185/$). Methodology: volume-weighted average between the entry bound (mana store) and exit bound (USDC prize drawings), weighted by June mana flow through each channel. Data pulled from the API 2026-07-01.
Entry: M100/$ stripe (no sales since ≥ Apr 1), M109/$ crypto. June volume 261k mana / $2,587 → blended M101/$.
Exit (USDC sweepstakes — the stronger exit): Drawing #30 ($1000, closes 8/15) has 1.21M tickets on the bonding curve → marginal price M0.221/ticket → naive spot rate M268/$. A marginal buyer today should discount for expected future dilution (+25–40k mana over 45 remaining days at the observed tail run-rate) → effective M290–307/$. Marginal beats pool-average as an "exchange rate as of [date]" — though neither is quite right given the first-click race the bonding curve creates. Reference points: #29 realized M139/$, May $100-dailies realized M241/$, charity #2 realized M179/$, charity #3 spot M146/$.
VWAP: entry 261k @ 101 + sweepstakes 232k @ 290–307 → ~191–198. Adding the charity channel (237k @ 146, donation-valued so a weaker bound) pulls to ~177–181. Rejected variants: realized-average exits give 119–151 (backward-looking, not a spot rate).
Caveats: #30's pool is whale-concentrated (top 3 = 62%) and front-loaded (67% of volume on day one), so the marginal price is noisier than the formula suggests. Big picture: lottery exits strengthened ~5–8× since market creation (charity implied M1153/$ in January → M140–300/$ now) as prize pools grew and participation thinned. Full analysis saved offline; happy to share details.
@EvanDaniel Oops. That was supposed to come from the bot account, not the main. Ah well, either is fine. I discussed the resolution with Fable; its analysis roughly matched mine, and it had more detail on the volume weighting than I'd bothered to collect.
@JussiVilleHeiskanen For the record, you are absolutely allowed to attempt to sway me on this or future similar markets! I appreciate it when people disclose conflicts of interest and intent, but all contributions are welcome.
Do you think you would also be inclined to pay less for tasks compensated in mana than dollars? Do you think the effect for you, personally, is roughly symmetric?
@EvanDaniel Hmm. I actually thought I was arguing counter to my holdings, in that my personal valuation would have been in the one bracket that I actually didn't hold one way or the other. My trading was influenced by things other than my personal views.
@EvanDaniel Not sure if I would be inclined to offer bounties except for things I wanted dearly, and then my desire would rule the price, rather than how much work I was asking for.
Current opinion:
You can buy mana at M120/$1 in the store, not including one-time discounts. (M120000 for $1000.)
One-time discounts would put it at M140000 for $1000.
The prize drawing curve is currently at 453.8 entries per M100, in a pool of 1203396 entries to win $1000. That's a marginal rate of M265/$1.
Presumably the true rate is between these two figures. If anyone has a reason it isn't, please speak up!
The prize drawing curve is currently at 453.8 entries per M100, in a pool of 1203396 entries to win $1000. That's a marginal rate of M265/$1.
I think this is not quite correct. The most a single entry can win is $500 and the same user can't win twice in the same drawing. So at the very best you have to say there are 1203396 entries to win $500 (but it's not actually $500, it's 500 or 250 or 100 or 100 or 50, with up to 5 chances).
I think this is strong evidence that users are willing to pay much more than 265 mana for $1, up to as much as 500 mana for $1. People are still regularly buying entries in this drawing.
----
Another possible angle:
For the last 10 daily $100 prize drawings, the total mana put into the machine was about 244,000 and the total prize payout was $1000, so you could look at that and say it established a regular, recurring exchange rate where site users as a whole gave 244 mana for 1 dollar.
That time period was the closest example of a repeating, regular scheme to turn mana into dollars. You could also make a calculation based on the final ticket sold each day, etc., but I like this big picture way of looking at it.
Another key differentiator here is that it was a winner-takes-all drawing so it was very clear that 1 entry was 1 chance to win exactly a certain amount.
It should also be noted: The current prize drawing was announced as "the last one for now", and this market is about the exchange rate at a single point in time. The current drawing is evidence of what is happening right now, while the past drawings should be considered background/foundational information but not critical to the current situation.
@Eliza Is the expected value of a single entry just total prizes / total entries? Regardless of how they split?
@EvanDaniel I guess the expectation for you declines slightly if you've bought a relevant fraction of the entire drawing? But not for a new user making a small purchase.
@EvanDaniel The cost of my entry depends on when I enter, but I still need to take into account how many total entries I think will be sold.
I don't have any direct experience with the prize drawing system but I did make large early ticket purchases in all three charity drawings. Each time I tried to estimate how many total tickets I thought would be sold, before making a purchase.
For example, on Charity Drawing #2 I estimated that it would sell "a bit fewer tickets" than the first one (say, 2 to 2.5 million rather than 3 million), and I was able to get 725,000 tickets for 100,000 mana to give $1000 to charity. So at that time my estimated rate was about 275-325 mana per USD of charity donation. In the end, after my initial decision, they changed the rules and added extra mana to the prize pool which made the rate look considerably worse in hindsight, but I could only go based on the information I had at the time.
For the third charity drawing I estimated that they would sell again sell "a bit fewer tickets" than the second one (say, 1.5-1.75 million rather than 2-2.5), and I got 705k tickets for 100k mana. This would be a little lower rate, like the 200-250 mana per dollar range.
@JussiVilleHeiskanen Absolutely! Right now there's an exchange rate with a spread, one side on the raffle and the other on purchases.
I don't want to commit to a plan to turn a spread into a point number without thinking about it carefully, but picking a result outside of the obvious spread would require a compelling argument. Some sort of volume weighting between the two trade modes might make sense, but I'm certainly open to other suggestions.
@EvanDaniel Yes, and somebody could snipe this market by manipulating the raffle of course if it even is stable as a feature.
@JussiVilleHeiskanen If I manage to disrupt the finance market by creating a derivative that's liquid enough I will laugh my ass off.
very much open to interpretation, but it's surprising/striking to me that the manifest tickets are not selling out particularly quickly.
this is people being offered a way to use mana at a 100:1 conversion rate, and people aren't rushing to take it. obviously, there are other logistical complications here—not everyone is going to manifest! but there are hundreds of people buying full-price manifest tickets, and if mana was worth vastly less than 100:1, you'd think there'd be more effort for people to transfer the mana to those who need to buy tix. presumably the main barrier is that it is against the rules to have someone pay you USD and send you the mana in return for them to buy a ticket... but the premise of a mana:USD exchange rate beyond the official one basically assumes the existence of such a black market, and it seems people aren't racing to cash out at 100:1.
(the other explanation is just that people aren't aware/aren't bothering to do it. but still, i'm surprised! in general i would say mana is not nearly worth 100:1, but this is the most direct exchange we have).
(and much weaker evidence is whatever the pace of sales is for the merch, although it's less straightforwardly clear that people would buy that without using mana—with manifest, we know lots of people are paying USD for tickets)
@Ziddletwix my guess is there are a lot of people who bought tickets before it was available through Mana purchases?
@Ziddletwix And the current prize drawing prices mana at about M$160/USDC, if I got the math right, if there aren't too many more ticket purchases. Presumably the 100 vs. 160 discrepancy isn't big enough to be a huge motivator.
@EvanDaniel obviously people don't value some aspect of the real money drawing the same, because the charity drawing is buying down past 300 at the same time....
@Eliza I hadn't even realized there was a charity drawing running in parallel. That's weird how different they are!
We are nearing the final hours of the first Mana Prize Drawing and entries are slowly trickling in still.
Some of the latest buyers have paid 100 mana for about 360 entries. I guess it's tricky to compute the exact EV on these without downloading the entire data set, since winners of higher prizes are not eligible for lower prizes, but if we just take the $500 first prize as a baseline:
360 entries out of 1778773 total entries for a chance of winning $500, seems like an expected value of about 10 USDC-cents? If we assume all the winners are just from single ticket buyers, the other prizes combined would amount to 4 cents, 2 cents, and 4x1 cent. So adding all that up maybe an EV of a hair over 20 USDC-cents for 100 mana.
That is filtering out many users who do not want USDC or do not want to cash out at this time, but if people thought this was a really good deal they would still be buying lots of shares.
Of course, earlier buyers bought in with a "better" rate but they had to have some guess of how many tickets would be sold in total.
I think I'll buy a little Yes in the 300-500 category.
@Eliza By comparison, the charity drawing was still selling 1000s of mana worth of tickets near the end, as far down as 1500 mana per EV-charity-dollar.
Setting up my own charity cashout at 250/$: https://manifold.markets/wasabipesto/how-much-will-i-donate-to-charity-a
@EvanDaniel I immediately took him up on the offer at 250 and then just now I remembered my target (posted in this thread) was 300-400. So there's another point of evidence!