XPY Flashstaking Formula | Options and Opportunities #XIOdebate

At the core of the flashstaking protocol is the XPY formula. It helps determine the universal interest rate across the entire ecosystem.

If too low, stakers may become disinterested in the protocol. If too high, we run into the obstacle of an unsustainable token supply.

In addition to the interest rate being high/low, the curve of the XPY is crucial in figuring out how much and long participants will stake.

Today, we are presenting two alternative options to the current formula to get feedback.

  1. Decreasing Linear (current)
  2. Sigmoid Function
  3. Bell Curve

Let’s get nerdy and debate the pros and cons of each! If you feel there is a fourth option that we don’t have that could be more efficient, please propose it!

I mentioned “XIO Feedback” in the video, but this will be an #XIOdebate social campaign. For extra credits, make sure to comment directly on youtube!

9 Likes

for extra credits please comment and like also directly on youtube: XPY Flashstaking Formula | Options and Opportunities #XIOdebate - YouTube

5 Likes

type 1

type 2:

type 3

2 Likes

I will not even begin to comment on options two and three, because while I do understand How they would work, I cannot say see any reason for options two and three.

Bonus question:
Portals v1 should be immutable.
Stability and certainty is vital to obtain investors’ initial trust.

If required, create new parallel Portals or alternately create an upgrade option as Zach proposed, although I do not know if that leaves more doors open.
I am would be interested to learn more about the differences between those two options.

1 Like

Type 3 would be a big improvement:

  1. Failed transactions when the rate goes up, possibly because of slippage, has been a big problem during peak staking times. The Purple option would be best to limit that.
  2. Whale holders dominate the crypto market, so it would be great if the small guy or late comer to crypto could have a fair opportunity to invest and stake in a new company.
  3. Being able to spread the higher % over time will be beneficial to a global audience of stakers because of time zones. (we all need good sleep, this is medical advice :stethoscope:)

Thank you for a fun competition and great community!

XID-f56A0
#XIOSocial

1 Like

I’m not the math guy but there’s something that aways pick my attention is how it will influence the price. We have 4 variables: timespan, amount of XIO staked, XIO price and interest. What if the protocol have an algo that burns more % XIO if the interest is low and “mining” more if the interest is high like a rebase mechanic’s but attached to to interest instead of price. It can happens at random hour daily or every time the interest high or low level. Please, some of the math guys could help me to see if this make any sense :slight_smile:

Bonus question: Newcomers will be more “into” adopt if we give them an stable ambience. If we need to change in future. The opt in idea is my take as well!

XID-77fd

The bell curve is very interesting to me. Let me explain in an example on the charts given (i know the numbers are not real but they show the example well)
With the first version we can assume that people will stake until the reward is too low and then they won’t add anymore money. For the sake of example lets say the number is 30% per year (I know that’s high). So in the first graph this would be equal to 40 (million/thousands whatever) XIO
With the third graph (blue line) this would be equal to 10-90 XIO. So up to 90 XIO wuld be staked. This means that over twice as much moeny will be staked so the value of XIO is higher.
So the question now is how do we get people to start staking? No one wants to start if the starting stake is less than 30% (in our example) so how do we get that first 10 (million?) XIO?
Maybe the treasury could keep some of its XIO staked?

Bonus question: probably immutable. If we have the opt in function it gives us the best of both worlds.

This is an excellent video,
I get the impression that the third option will encourage early stakers to stake for a shorter time, hoping that others then stake to raise the interest rate and when their short stake ends, they can restake for a better rate. Assuming you can get in at the time of high interest, you would then be incentivized to stake a loner time from there.
Option 1 - I like the simplicity factor but it does favour whales if they can get in early.
Option 2 - does a nice job of negating the extremes of very high or very low interest rates.

Assuming we dont go for option three, I think it might be an idea to reduce the starting point of interest from close to 50 percent and that way the rate of change on the way to zero would be less dramatic.
I think the flashstaking competitions are the perfect opportunity to try out these different formulas, and we should definitely do that, there is some interesting learnings to take away from there.
Overall, i dont feel qualified to say one is better than another, they all kind of play into a system which people will try to learn and figure out how to game for best returns, So we should look to see what formula will create the most XIO staked for the most time, and the best distribution of rewards.

From a general interest point of view, i think option three is pretty cool and worth exploring as it has something in common with how traders try to time their market trading by going short or long on an asset, so you could see some interesting activity using that formula.

Bonus question is a good one, i’m going to need more time to answer that one.

XID-AF60F

I know that option 1 works having seen it in action during the testing period with the stimulus acting as new tokens being introduced over time. The only issue is the large failure rate as more people tried to stake at the same time causing large amounts of slippage. This is most likely due to the planned nature of the stimulus whereas in real life there would never be a fixed period where the interest rate would spike repeatedly every four hours, meaning a more steady flow of stakes taking place.
The second option sounds like a good way to counter the rapid decline in the rate following the first few stakes. I’m guessing it would hover around a certain level (15/25% mark) and would meet larger resistance to push higher or lower than just a few large stakes being started or ended.
Would be interesting to have a smaller trial using that formula and see what happened to the % rates.
Option 3 doesn’t sound great to me as you could have people waiting to stake until others do which doesn’t help the platform.
XID: 61f1

The maths involved in here is a little overwhelming haha but based on the fundamentals of the various formulas Dash talked about, I think I would select 2 and 3 for the next stage of testing to further weigh their pros and cons and make a final decision on the best formular to use. The fact that whale dominance, time of stake and variable XPYs are accounted for in the later curves also influenced my interest in them. It fine tunes the whole protocol further.

One other concern which was raised was how do upgrade after launch when the need arises. Governance voting isn’t something I would vouch for at this early stage for the project. For now I would suggest we give full concentration to the staking structure development. Just like uniswap we could go for an initial launch without a governance system and later when we’ve fully developed everything we can then go in for it. In view of this when the need arises for say a formular upgrade on the XPY, it would be best to build a whole new version than use a voting system to make the changes at least for the early stages of the project. Just my suggestion.

First of all, congrats for editing the video on the fly - that is pretty impressive and will be very handy in getting information out quicker.

Comparing the hyperbolic tangent function (tan(x)) and the liner function we are currently using where the fewer people staking the higher the reward. I am not a fan of using the tan function for the XPY - this VERY heavily benefits the very first stakers as opposed to the first bunch of stakers. I think it should be a linear gradient function similar to what we have at the moment (the chart showing the triangle). I have revisited this after my initial comments on Twitter and have now had a chance to look at the graphs provided in detail. It looks like we are not using the same sort of formula as I had initially thought. The additional reward for being first is not as great as I initially imagined - that being said, I don’t think there is any need for this - why are we trying to reward the first stakers more so than say the next 5?

If we are looking to ensure the very first stakers benefit slightly more then we could just increase or decrease the slope in the linear function. (fewer people staking the more you get). The third or fourth option you suggested (from yearn founder) also has me a little worried - would this not stop people from staking until it reaches the ideal return? I think this is something that we should keep in mind.

In terms of the formula being used, I really see no actual benefit to moving away from the linear equation being used at the moment - I am not sure what problem this is looking to achieve. The harder we make the maths behind the flash staking to understand the more we potentially alienate the users.

Bonus question: I really liked the idea of using a combination of the two. As I mentioned on Twitter there must be some balance between immutability and flexibility. There was a mention of being able to opt in for a new formula - I really like this idea, its very similar to Uniswap V1 and Uniswap V2.

1 Like

I can’t see those who jump in early would have such an advantage. It could also detrimental as those who didn’t get in early could think it’s too late and not really worth it anymore. So as far as I’m concerned type 2 or especially 3 seems better

Bonus question: Trust is important. But if the formula would appear to be inadequate and no change is made, it will dwindle. Stability is important, but it must leave space for changes if need be. The opt in option could be good, but only if it doesn’t end up dividing the community. If there’s a risk of dividing it, it would be better to chose a different way. I think it’s mostly a matter of communication and transparency. Giving a clear frame so people know what to expect, but specifying the formula could be changed after a given period of time if it appears necessary for instance.

Well…This one is very interesting …on the first thoughts I would definitely want a function which doesnt have a definite advantake to early stakers…Type 3 is the function for me as that will help

  1. Allows the person who missed a few days to still stake and get somewhere near highest yield
  2. As the stakers dont know where exactly the top is, people will be incentivised to stake till they find the optimal XPY…at which stake they have a even bigger incentive to stake more.
  3. I feel the purple line is the better one as that allows a more flatter rate throughout the period.

I honestly feel that having a “certainity” around the returns and the function is good as that builds confidence in the protocol from long term hodlers(like the XPY builds long term thinking ).

I would not want to tamper the protocol unless something really is broken in the system.

I think bell curve is interesting but if we put that math function in the smart contract will it become heavy? and with high fees? XID-F524

Hi Mr. Dash

Nice informative video, it’s so strong for the community that you take questions of this magnitude with your Citizens.

The bonus question first: Introducing uncertainty will not be the optimal way of encouraging people to stake. But as you say if there is a better way it could be important to have the opportunity to make a change. In my opinion it should be by vote, the more XIO you own the more voting power you have. It’s important by my opinion that the majority should be at least 75% to make a big change like this.

By the look of the 3 formula I’m would prefer the Bell Curve if you some how can introduce a start percentage of like 10% to encourage early adopters.

XID-AAA2E

#XIOdebate

It’s great that the team are seriously looking at this as it is such a core component of the platform. I think the linear formula is super dangerous as we’ve experienced in the comp, as whales will always dominate. I know that’s good in terms of increasing volume and liquidity, but it will impact mainstream adoption by the masses. XIO should be an every day person’s tool to grow wealth and participate in projects, not for whales to hoard tokens then potentially pump and dump the market, to the detriment of everyone.

I like the sigmoid curve because it provides fairness and consistency. When it comes to building trust, esp with new users, this is so important. Nothing worse than be introduced to a new platform and you feel disadvantaged in your overall experience. To create sticky users, the platform should be fair and generous to all, and I think this option is best to provide that. The bell curve, whilst logical, will be to the advantage of whales who can build bots to take advantage of the peaks, and use volume dominance to influence their gains. It’s almost like the linear method, except it’s more to do with timing… and it would be easy to create bots to take advantage (I think).

Re bonus, I believe in flexibility as the world and crypto and everything else are always changing. Users expect and understand change. I absolutely think there should be a governance model to ensure consensus, and I like the idea of v1 v2 etc to control the experience, so those who prefer one method can stick with it. The more flexible (yet simple) the platform is, the more likeability and stickiness it’ll create.

1 Like

I agree with 3rd point, it will be really beneficial to have similar opportunities in different time zones.

Linear option, we saw in action and it was favorable to early stakers (timezone cry baby here), Sigmoid function made sense at first but it is way too similar to linear and is sort of exponential. Bell curve makes sense in terms of xpy, it might also work random because people will try to unstake and stake to find the top and that could work in favour.
Nothing good ever came from splitting chains, It makes logical sense but in the end community is always divided, and one thing I love about XIO is the community!

As another citizen has said, I vote for a simple linear function. Simplicity is key in this space, especially this year since a lot of new users will probably enter the crypto space. A linear decreasing function will ensure the long term sustainability of the staking portals.

As for the bonus question, I stick to simplicity as well. Immutable portals are important for citizens and investors to plan their investments and moves

Edit: i thought more about the possibility of whales suppressing the interest rate, if we use the linear formula. Well, now it’s a tough one :thinking:

Another edit:
Watching the three curves images here on the xio forum, I really can’t wrap my head around the third curve type. If there are a few people staking, it should increase the pressure to stake more, but if the number of users stays low, we all suffer from low % as a result :thinking:

The Linear curve is generally utilized and tried on different occasions in past, it works honorably and it’s fundamental and easy to comprehend.
Sigmoid curve appears as though it gives higher rate to longer period however don’t know a lot about it and need to look other model how it will profit our ecosystem and help stakers to stake for long haul.
The third bell curve proposed by Jorno seems as though it will be hang tight for the best rather then “stake and don’t stress” since it would keep people from marking until and except if it arrives at the pick for maximum return.
I truly observe no real advantage to moving ceaselessly from the linear curve being utilized right now as it is working perfectly and if we choose other do we know it would give better result than the one right now?? But its always good to have couple of backup options which will always benefits the protocol.
Bonus question: There is likely going to be a need to make changes as crypto is regularly developing space I genuinely adored using a mix of the permanence and adaptability with new redesigned concept and formulas.
XID-A31e