Value of NMR token - incentives?


What is the incentive to hold/trade a NMR token?

In reading through the relevant docs this doesn’t seem that clear. A “data scientist” gets paid in NMR, then goes to cash out at one of the crypto exchanges - but why would a trader want to own it? Why would it have value?

Seems the market is uncertain about this too so NMR is not exactly stable.


That price decline is into rising volume. Someone is selling. It’s either Numerai themselves raising capital through the ICO or one of the large winners like Xirax or WSW, who is sitting on thousands of the currency, locking in value.

Interesting to note that total supply is 2.2m and we know around 1m has been allocated to participants, give or take. That would give Numerai around a million for raising capital or $22m USD at current price levels. Not a bad round of funding if the price stabilizes. Unfortunately, the market is littered with ICO’s that are selling for pennies. There tends to be very little interest in the currencies once they hit the open market as there is no direct tie-back to the firms value. The market also has to value total Nuemeraire available and we know there are around 21 million that can find their way into the market. That’s like printing money and will create a deflationary pressure on the currency if there is no additional interest or need for it.

This printing risk will persist as there is no clear cap in place, there is nothing to say Numerai won’t and can’t print more beyond that. If values fall below a dollar and they find themselves burning through cash, they may need to adjust their cap, maybe $30m becomes the total figure. That will create more pressure.

We can’t know for certain how any of this plays out, only time will tell.


Let me rephrase and be more direct - why would anyone buy the Numerai token?

If I understand correctly there are 2 types of users coming in contact with NRM:
a) data scientists
b) exchange traders

Data scientists just want to unload NMR that they won and get paid - they basically never buy NRM tokens with their own fiat/money (I know they could - but they most probably don’t). Exchange traders could buy NMR - but why? Just to support the Numerai fund and get nothing out of it? They’re not investors, have no voting privileges, no dividends being paid out - the tokens don’t reflect the value of the fund, they’re not even deflationary (like some tokens that use a part of the profit to burn tokens, thus making them scarce over time and more valuable).

It seems like a flawed model - Numerai management/investors are happy because they their platform grows, data scientists are happy because they get paid for their work, but what do people that purchase NRM on an exchange get? I’ve read through the whitepaper/wiki/articles - there is not even a hint of promise of what that might be. Even the bad ICOs out there try to promise the people purchasing their tokens at least something for their money.

The only reason this hasn’t imploded yet is because there are so many cryptos now and so many newbie investors that just buy blindly that nobody noticed or cared yet. But its going to get noticed. I’ve watched it get listed yesterday on Bittrex exchange and there were 70BTC committed on the bid/buy side and 100BTC on ask/sell - not a bad spread. Today there are 150BTC bids from people buying and 1100BTC of people selling. Definitely not just 2-3 people selling.


Holding NMR has an obvious benefit (to both data scientists and traders/investors):
Data scientists have a stronger incentive to hold it as it represents a pay-out from the success of the fund i.e. given that the data scientists presume the fund will be profitable, then the cashing out NMR has a severe opportunity cost: forgone future returns. Analogous to receiving shares in a company you work for as a year-end bonus: yes you can cash them out now for money, but if you believe the company will grow then it’s optimal to hold onto the shares. Early sellers may simply misunderstand the economic incentive model. If one assumes that the collective insight of 12000 data scientists can be successful – then the logical decision would be to hold NMR.

12000 data scientists contributing ideas at the stake of their ‘salary’ (remember they have to stake means successful algorithms are rewarded while unsuccessful algorithms have their NMR ‘burnt’ i.e. deflationary and supportive of a higher NMR price. The requirment for NMR to stake is also a demand driver. As I’m sure you can tell, this is also a postive-reinforcing cycle i.e. higher NMR price is a higher incentive for data scientists.


Hi Diageo, thanks for offering an explanation however I’m not clear. Secondly, from what I’ve read so far, I’m fascinated. Maybe I’m missing the obvious and I’m being lazy but I haven’t been able to find a direct correlation between being an owner of NMR, the coin, and being an owner of the underlying assets of the fund that all the data scientists are contributing their ideas to or somehow being an owner of the returns of the fund. If NMR coin owners are owners of the fund, is there a way to know how much each NMR represents in dollar terms or ETH terms. What is the cash value of an NMR and how much would I be paying for the “secret sauce” so to speak?


Ned some of what you’re asking and I’ve also eluded to isn’t limited to Numerai or its skyrocketing currency (currently $95 a numeraire). We could ask the same of Bitcoin, or Ethereum. Honestly, value questions can also be asked of stocks and exchange rates to a certain extent. What we often end up witnessing in the markets is the phenomenon of “interest”. People get interested in this thing over the other and they gravitate towards it and that creates demand.

Numeraire may succeed simply because it has exposure that other’s don’t. That interest and exposure may create a liquid/actively traded token (that’s hugely important) that is targeted by Crypto investors and holders for diversification. Add to liquidity a decent buy-in price vs Bitcoin and Ethereum and that helps make the diversification argument even better. I don’t know, thinking out loud and speculating… high broad strokes.

Having just a few of these in my pocket right now is making me pretty happy (wish I still had access to my old FXanalytics account… but I think I torpedoed it, meh). That I can’t lie about. I also have to believe given the payout structure of Numeraire in the competition and the current price levels, it will absolutely grab the attention of some very capable Data-Scientists (glad I got mine when I could lol). We are now getting into a reward system that is going to hit “lucrative” levels. People who were casual about the competition or brushed it off because it didn’t pay well enough or offer an incentive will now be completely “incentivized”. This also makes the “anonymous” Data Scientist, more meaningful. If a quant at another fund or elsewhere really shouldn’t be moonlighting at Numerai, who’s to say she is… The big picture is taking shape. Pretty cool stuff.


Thanks Objec,

I appreciate that gold, stock, bonds and cryptos all are valued with a myriad of components and maybe emotion is one of them. I would like to buy a little just to be a cheerleader in this wonderful project and reward a fine data scientist for their predictive abilities. However, I would be more likely to buy more if I was actually buying assets in the portfolio and my return was tied to the funds Longterm performance.

When you bought NMR, did you think you were investing capital in the fund and were going to be rewarded if the fund performed well? Or did you buy because of some other reason than the next person may buy NMR for more than you paid? I’m not trying to be a wise guy, just trying to see if NMR is tied to the fund’s performance or its AUM, assets under management.




Ned to be honest, I’m one of the fortunate ones that has actually earned some from participating in the DS competition. My cost was sweat equity and some computational resources.

From the perspective of an investor I completely understand your questions about buying and holding and it’s hard for me to point you to a direct link back to the firms value. As I’m no longer a CTA and don’t really know how well this kind of speculation would fit into your broader portfolio (risk profile, correlation to other assets etc.), I can’t really give you advice one way or the other.

I would say this, I understand your hesitation and I think it has merit, these were the same reasons I never put $ into Bitcoin. Now I wish I had risked a little… But more often than not I’ve been right and probably what I had made on BC would have been lost many times over on the ones where I was correct about value.

Hindsight is always 20/20 and it’s easy to see the ones where you got it wrong because they will always be touted and flaunted… If that link to value is important, if that is a cornerstone of your portfolio maybe it’s time for some uncorrelated thinking or maybe you need to hold true to that. You have to weigh the risk: n(sleeplessNights x opportunityLost) vs n(sleeplessNights x moneyLost). Which one will keep you awake more is typically the one I try to avoid.

If you do jump in, given your hesitation about underlying value, it’s probably a good idea to keep your exposure small, think of it as a Vegas fund. You’re going to throw a little on the table and if you win you win if you lose you lose either way the impact it has one way or the other isn’t huge. At best you get some diversification at worse you lose some beer money.

[sidenote: In typical Analyst fashion I think I’ve said a lot while saying absolutely nothing at all… I should be on CNBC]



Again I appreciate yout time and response. A bit about my past, I traded stocks for a big international bank and on the floor of the NYSE. I’ve owned internet dot com darlings and utility stocks. The former I knew that I was buying dream and the latter I was buying a slow growth dividend stock. I didn’t know the outcomes beforehand and the risks were wildly different but each time I knew what I was ‘betting’ on. I don’t mind risk, I’d just like to know what the ‘bet’ is.

Am I hoping no one has any good ideas so that no NMR will be awarded, so that a dearth of selling will create a run up in prices OR do I want the fund to have super sized gains and that will be reflected in the price of NMR?



If I was still on the East Coast working with Tuono I’d say meet me at Delmonico’s and we’ll figure it out. Between that and “good-ole day’s” stories I’m sure we could hash out an answer.

The bet here is hazy because the link to the issuer’s value is unclear, if it exists at all… Unless of course the issuer has a vested interest in pushing the value of the currency higher to attract talent. If a DS were to pull in several hundred or more of these a month it justifies their efforts. The link to value may be an intentional one if not a fundamental one… Based on current price levels I expect the pool of competitors to get deep quick. That will of course give the firm the best possible shot at making this a reality.

Looking at SEC filings it looks like investment on the fund side has been a little slow, people may be waiting for proof-of-concept or returns in year one were soft. Richard has a hedge fund background and I’ve watched him make some really critical and very smart adjustments over the last 18 months that I’ve been around. If it can be done, I think he can do it. That’s not smoke and I don’t have any connection to the company, just an observation. He’s also plugged in some really bright and creative talent. I think they understand (or are figuring out) what it’s going to take to acquire the talent in this framework that will give them the best opportunity to generate returns.

This is the Uber of hedge fund structures. They’ve basically built out a mass amount of computational power and man hours for a fraction of the cost of doing it internally. Now they are working on finding the right incentives to pull in the very best talent.

If Refco hadn’t run off with my money/company in 06 and put me in the poor house, I’d be in this thing on the back-end as an investor, the front with the fund and in Numeraire. I think somewhere in that mix there is some very real potential, in IP, outright returns as well as token appreciation. Obviously I wouldn’t bet the farm, but I’d def have a few of the chickens in there.

JP .


Thanks so for the insight! I am going to throw something at it. I just don’t know if I should try to even evaluate it and make my own guess as to where it can go.


24 hours or so ago it was in the mid 20’s so the risk was a little more asymmetrical. With it being based on Ethereum, short of really finding a link back to the company in value, it’s hard for me to see how it eclipses that in price. Not saying it can’t just not sure how or why. If we speculated (Ethereum minus some percentage) we could be sitting between a third and a half of potential value… Just throwing it a wall at this point. Maybe buy on the dips??? The last 24 hours definitely ate up some of the upside but that doesn’t mean we’ve found the top either.

Good luck, I hope it pans out well for all of us involved.



Great questions, you highlight a key concern with the model: hasn’t (to my knowledge) given clarity on a major issue: Is their allocation of NMR tied to hedge fund returns. The reason why this is so important is simply because NMR represents the compensation of Data scientists - so investors/speculators in the market who are choosing whether or not to buy NMR (which pushes up the price) need to be assured that NMR is being paid in relation to the success of the hedge fund.

In my opinion NMR should be tied to the performance of the hedge fund, for the simple reason that the Data scientists’ work is effectively creating the returns (therefore they should be compensated accordingly and in proportion to the extent of outperformance - this is the way ‘conventional’ fund compensate their analysts).

Clarity on the pay-out formula would also provide the basis for creating a more realistic valuation model of NMR i.e. one could assess, given the return of the hedge fund, how valuable the contributions of the Data scientists have been - and therefore assess the appropriate pay-out of NMR vs NMR supply.

My post above is based on a key assumption (which may prove to be wrong): I believe is incentivised to have NMR price increase (as higher NMR price effectively represents a higher pay-out to Data scientists, which in turn should create demand / attract more contributions, from more scientists).

If however my assumption is wrong, and aren’t paying out a ‘fair’ amount of NMR (i.e. paying out much less that the returns earned in the fund) - then I believe there is an argument to be made that the ‘market’ (traders/investors in NMR on exchanges) won’t believe the value of NMR 9because of the fundamental risk that NMR isn’t linked to the hedge fund returns / Data scientists work) which would cause significant downside pressure to the demand of NMR. Said simply, if can’t clarify that they are fairly paying their Data scientists, then I believe it’s only a matter of time before the market realises this and discounts the price. On a personal level I’m sure wouldn’t let this happen, though that is a subjective opinion.

  • please see my reply to this post. A lot of the thoughts here are incorrect, my apologies.

The white paper states [quote]The Ethereum smart contract dictates there will never be more than 21 million Numeraire minted[/quote]

You can see this cap in the actual contract code [quote] // Cap the total supply and the weekly supply
uint256 public supply_cap = 21000000e18; // 21 million[/quote]

I’m trying to interpret your printing risk comment and see three possibilities (in order of probability):

1. You are referring to numerai printing new NMR after burning it during a previous tournament
This seems likely to me. Burning NMR every tournament could make it scarce. I think nothing prevents numerai from printing up to 21m NMR. If it is profitable to do so, why wouldn’t they? Does this create a perverse incentive for numerai to steal NMR from participants by submitting false scores to the smart contract.

2. You think there is no real cap on the NMR.
If this is the case then please see the smart contract code indicates a hard cap at 21m.

3. You think a new numerai will somehow circumvent the smart contract cap.
I’m not sure how they would do this. Maybe print a new coin or something. This seems unlikely to me

When it comes to the value of NMR I think this is proportional to the prize pool and the probability of submitting a winning algorithm. A few scenarios which I haven’t spent much time thinking about (maybe you can improve them)

trivial success P(success) = 1.0
If it is trivial to submit an algorithm with logloss < ln(0.5) then all players submit an infinite confidence and stake all their numeraire. In this case the payout to each player is 0 (assuming multiple participants).

trivial failure P(success) = 0.0
In this case no players ever submit an entry and numeraire value is 0.

uniform random success P(success) = U(0,1)
I take this to mean that probability of success is uniformly distributed and not dependent on prior testing. In this case the tournament becomes a lottery where each submitter has an equal chance of winning. This is where it gets interesting and I don’t have time to work out the math.

The white paper addresses this question of value by considering confidence separately for each participant. They say that for a player with probability of success = 1 the value of numeraire is the net present value of all future stake payouts (confidence / stake). I think this ignores the competition aspect of the tournament but I’m not sure (i.e. with 1 participant who has P=1 this statement would be the net present value of all future tournament payouts, but with 2 omniscient participants there is a tragedy of the commons problem where each player must set confidence to infinite.)

I think data scientists are also making NMR during tournaments? This wasn’t clear to me in the white paper, but some emails made it seem so.


Hmm… It looks like a lot of these comments are no longer valid b/c of changes in the tournament structure posted at

-edit1 Also it makes senses to look at the value as a function of the confidence/stake/payout of participants in previous tournaments.

-edit2 If it wasn’t clear, I have no idea what I’m talking about and just started reading about NMR, hopefully I don’t give anybody a different impression. It looks like a lot of what I said is wrong when it comes to the value analysis. Its possible what I said about NMR printing is correct, but I doubt it.