$1 million weekly prizes & POOL distribution

41

It’s trader’s intuition. I can give you some rough data that gives me these strong hunches about the dynamics at play. Your predictive models are only as good as the assumptions you make so I will give you ones I have made and give some rough data that I think supports them.

Assumptions:

1). The vast majority of the POOL being sold each day is derived from the liquidity mining associated with USDC and DAI LMs.
2). If we reduce this reward drastically it will have a profound upwards effect on the price.
3). The upward effect will compensate to maintain the current APYs on the pools.
4). We will be getting more bang for our buck; hence diminishing returns will decline

So for starters, take the top list of addresses displayed in pooltogether for the biggest players in the USDC and DAI pools. See which ones are selling and which ones aren’t. Find the proportion and apply that profile to the entire profile of players. That seems reasonable to me. I found that 86% of DAI (skewed heavily by the 35mm depositor) and 63% of USDC POOL that’s dripped is being sold on the market. That means 0.632295.1 + 0.862295.1 = 3419.699 POOL @ 19.4 = 66,342 USD worth of POOL is on average sold each day.

The average volume on Uniswap per day over the last month is roughly 500k doing a rough eye test. That means each time this is sold and bought equates for 66,342*4 = 265,368.64 USD worth of that volume. That’s 53% worth of the volume just to maintain the price at a certain level where buyers and sellers are cancelling each other out. This doesn’t account for trading bots that are doing trading strategies or front running bots or arbitrage / rebalancing between sushiswap and other short term activities from traders, so that is a massive proportion of the volume imo. I know there are other markets listed on coingecko, but nearly all of that is fake. The real volume is on uniswap and some is on sushiswap so maybe it would be fair to have included that volume too. Anyway, back to this rough looking model.

Lets paint a hypothetical model:

You could model this, take the uniswap price action over the last 14 days or longer and assume that the liquidity miners that are selling are doing so with a reduced reward. So for example, if we cut the rewards in half farmer 1 is selling 100 POOL instead of 200 POOL. Let’s also assume that buyers behaviour does not change.You can then find out what the price of POOL would have been. Again, because all I ever do is ballpark, instead of 3419 POOL been sold to the market, what spread markup is there for only selling 1709.5. I get 1.69%. So each day, the market is 1.69% higher. Over the 2 week period, instead of the price being $19.4 2 weeks ago, it would be 19.41.0169^14 = $24.52 and 19.41.0169^30 = $32.07 after 1 month.

Of course markets don’t work like this, when the price goes up some buyers down at $19.4 will get put off buying and the buying power dissipates you could argue. You could also argue that they were accumulating down there and end up fomo’ing in once it starts rising believing they won’t be able to buy down at $19.4 again. And in reality a market doesn’t just go up in a straight line. So there are many issues with this simple model, but it does get across that a reduction in supply does lead to a significant change to the market dynamics over time. The more important question I ask is, how does that affect the mentality of the buyers in the market? We know how it affects the sellers; it doesn’t they are still selling every few days.

And finally I’ll just add, in order to keep the APY I listed above constant if we reduced the rewards in half the price would need to be $40.84.

Current: 2295.9x(20.42) / 79,779,971x365x100 = 21.449032% APY

Halving the supply and maintaining the APY
21.44903279,779,971 / (365100*(2295.9/2)) = $40.84

Certainly if I saw a signal that the rewards were drastically coming down I would be revaluing what I thought the value of POOL was on the market and I’m sure others would too. It’s clear to me, given how many times POOL has got supported in this $16-$20 price band, that the buying pressure down at $19.4 would overtime overwhelm the sellers.

5 Likes
42

Aside from all the points made above, I wanted to pitch the idea to combine Dai and USD prize pools. They are both pegged to 1 USD. If one goal is to advertise a larger prize, why not allow these pools to earn interest for a combined prize?

4 Likes
43

Alright, great discussion here! I’m going to try and summarize a few take-aways from this thread and next steps.

General Consensus:
It seems there is strong agreement on the following items.

  • POOL rewards should be significantly reduced and extended
  • LP incentives should be increased
  • Too early to commit to exact numbers for one year. We will target distribution schedule for one quarter with intention to go longer but rates may adjust after the quarter
  • We will favor USDC prize pool over all others, the core idea being that we want to have a single prize pool that is the default for new users to join and has the largest headline prize but still support prize pools with other asset types / yield sources
  • We will have a separate discussion on adjusting reserve rates
  • General consensus was very positive towards splitting POOL between sponsorship and depositors but given that Pods are coming online this week and this is already a big change we will not do that initially.

Next Steps:
I’ll use the above assumptions to craft a PTIP that I plan to post tomorrow. We should then start voting now later than Wednesday. The main thing to land is just exactly what POOL distribution rates we want to use. I’m going to model a few scenarios on what we should expect to see with different rates and propose something aligned with what has been discussed above.

Tagging: @ageless @RegisIsland @Torgin @Brendan @TheRealTuna as you were primary contributors on this thread.

5 Likes
44

That’s a great summary. A reduction in scope will make it much easier to reach consensus and see the impact from the changes.

2 Likes
45

I have a last minute contribution to the discussion above.

It seems like there’s been a lot of focus around attracting funds into the USDC pool to drive up the headline weekly winnings to $1m for publicity. I would assume that the $1m part of this plan is more important than the weekly cadence. Therefore we should consider reducing the cadence substantially to say 1 winner every 2 months for the USDC pool. For variety, the dai pool can maintain its weekly cadence and we leave the USDT pool at a daily cadence. This approach effectively gets us what we want without major changes/improvements/additions to the code. The assumption is that as the winnings grow, more people will be attracted to the pool and we can get to $1m faster than the ~13 weeks you’ll need if you assume the current user base stays constant.

Alternatively, we could also set the USDC pool to only declare a winner once the winnings get to $1M+. This improves the earlier suggestion by allowing for the possibility of really fast growth to shorten the wait time. The ‘+’ gives us some time after we hit $1M to run around and do some marketing. As usual, probability of winning will be a product of $ invested and time spent in the pool which creates an incentive for early depositors.

For this to work effectively, I side with the ideas to drop the reserve rate temporarily and focus on user growth in the short term while heavily guarding against ‘bad inflation’.

Edit: Added the $1M+ modification

2 Likes
46

I actually kind of love this idea. Yeah, less prizes in total, but if the driving force to growth is ‘headline prizes’ then this would surely do it.

In general, im kind of resigned to the fact that the probability of a win for me in any of the stablecoin pools is so small as to be basically not going to happen. The whales are only there for the liquidity mining, i dont think they are massively concerned about the winnings. They are a nice bonus when they happen for them. We can leverage that by issuing prizes less and pumping that sweet headline prize amount.

2 Likes
47

Wow, this king did the math. Love it haha.

2 Likes