SGB Reward Rate - what it means

[Seoulite]

1 hour ago, BillyOckham said:

Well I could be wrong but I think it’s that each provider creates a price reading every 4 minutes (??) and then at the end of that round only some (half??) of them get the reward (according to their vote power).

So if they are good at getting it “right” then they get more reward wins.  So  “overall” is their wins averaged across all time, and “this epoch” is just the current one.

 

That makes sense for the reward rate but the voting power number also changes if you click between overall and this epoch. Surely those numbers should be the same? Or is it counting the tokens delegated after the epoch has started?

[BillyOckham]

Just now, Seoulite said:

That makes sense for the reward rate but the voting power number also changes if you click between overall and this epoch. Surely those numbers should be the same? Or is it counting the tokens delegated after the epoch has started?

That’s a pretty much live look at delegations for the current one.  It changes as people migrate.  The overall is likely an average for the whole period.  I’m just guessing though.

[jbjnr]

3 hours ago, Seoulite said:

So it now says A FTSO is over the 10% threshold

As the others have mentioned, votes/rewards are capped. But bear in mind that the voting epoch started on Sat, so new votes going to A-FTSO shouldn't (if we're to believe the rules that nobody really explains) make any difference until the next epoch starts. I myself delegated to #2 and #3 on the list this time because I was worried that millions might pick A-FTSO at the last minute and go over the threshold before the epoch started.

[Not wanting to be too gloaty or anything, but "my sell the lot and earn % on stuff I don't own" strategy is going rather well, getting almost 1.8% on something I already sold, and at the same time, got 12% (so far?) on the stuff I bought. I'm giving this a trip-advisor * * * * * 5 star "must try" rating ].  

edit: change 2.5% to 1.8% because I did it wrong 

Edited October 10, 2021 by jbjnr

[Tull]

16 hours ago, BillyOckham said:

Yes I think that is the correct interpretation.  
 

So if the reward per 100 for that particular FTSO was 2 but only if they have less than maximum voting power, it will be less if oversubscribed.

Those who have delegated to such an “oversubscribed” FTSO will actually only get say 1.9 reward.  The delegator is rewarded according to vote power but only up to 10%.   After that it’s not getting more reward but does have more accounts getting a split.

Is there any way to figure out exactly how much the rewards would be reduced based on the amount an FTSO is over-delegated beyond the 10% limit? 

[BillyOckham]

1 hour ago, Tull said:

Is there any way to figure out exactly how much the rewards would be reduced based on the amount an FTSO is over-delegated beyond the 10% limit? 

Yeah I think it’s just the ratio of ‘over subscribed number / max subscribed number’.

But unless it’s a big over subscription it won’t ultimately be a big positional effect I think.   For instance 1% over (eg 11%) is a ten percent reduction (if I’m correct).  But the over-subscribed ones are likely to also be the high performers.  So losing 10% of a twice as good as most performers’ rewards is not a bad position to be in.

Its better if not oversubscribed but I don’t think it’s the end of the world.  

I think most people are worrying about minutiae that don’t matter at our scale.  Chasing the highest performer is possibly a risk.  But at the end of the day the rewards are pretty good no matter what you’ve done.  Just make sure you compound them and you’ll be doing fine.

 

[ftso_au]

On 10/10/2021 at 5:13 PM, Seoulite said:

That makes sense for the reward rate but the voting power number also changes if you click between overall and this epoch. Surely those numbers should be the same? Or is it counting the tokens delegated after the epoch has started?

The overall is the current amount delegated to that provider now. The overall number will become the vote power for that epoch when the voting lock happens.

Example, we're currently on 92,261,963, if we stay at that number come the voting lock on Thursday, then that will be our vote power for the epoch. 

Alex has rocketed past the cap so far, it's likely he'll keep climbing, then he'll be locked away over cap and the rewards diluted for the coming epoch.

[ftso_au]

10 hours ago, Tull said:

Is there any way to figure out exactly how much the rewards would be reduced based on the amount an FTSO is over-delegated beyond the 10% limit? 

There's no way and as suggested by @BillyOckham it's not really worth worrying about unless someone is grossly over the cap.

The other thing to consider is there's no guarantees for the reward rate each week.. While excelling at the moment, it's a possibility that his results could decline, see Optakt and AFOracle, both had much better results last epoch.

If his reward rate dropped by ~30%, and he was over cap by ~20%, then the weekly rewards would be much different.

Still, with epochs only a week long, it's not the end of the world, is it?

P.S there's a few new Signal Providers ready to go the next epoch, so they might shake things up.

[BillyOckham]

7 minutes ago, FTSO_AU said:

P.S there's a few new Signal Providers ready to go the next epoch, so they might shake things up.

I’m sure you deserve better results than you’ve so far had.  Do you know how or why some have so far been higher?

See the thing is that if the differences were small you would see some random variation in leadership.  But it appears fairly stable so far so that suggests that there is a systemic difference of some kind.

I can appreciate that in this competitive system you may not wish to discuss these issues (F1 teams don’t normally discuss technical advances until they are old hat) but are you able to speak about the why of this seeming performance difference?

Thanks for your presence and work, and I still support you,  but I will admit I changed ratio from you getting the lions share,  to a more even split.  Sorry.   

[Seoulite]

22 hours ago, BillyOckham said:

But unless it’s a big over subscription it won’t ultimately be a big positional effect I think.   For instance 1% over (eg 11%) is a ten percent reduction (if I’m correct).  But the over-subscribed ones are likely to also be the high performers.  So losing 10% of a twice as good as most performers’ rewards is not a bad position to be in.

Its better if not oversubscribed but I don’t think it’s the end of the world.  

I think most people are worrying about minutiae that don’t matter at our scale.

This has proven to be right. It seems the proportions have been evened out. Likely a small number of large holders moving millions of tokens, mostly into bifrost and scandinodes it seems.

[ftso_au]

19 hours ago, BillyOckham said:

Do you know how or why some have so far been higher?

That's the big question. It all comes down to being in the median price range, they've managed to do it more than us.

Interestingly, Bifrost's rewards (at the moment) are much lower than previous epochs, so despite having the most vote power, they're not "performing" as well as before.

Edited October 12, 2021 by FTSO_AU

[ftso_au]

18 hours ago, BillyOckham said:

Do you know how or why some have so far been higher?

That's the big question. It all comes down to being in the median price range, they've managed to do it more than us.

Interestingly, Bifrost's rewards (at the moment) are much lower than previous epochs, so despite having the most vote power, they're not "performing" as well before.

[ftso_au]

6 hours ago, Seoulite said:

This has proven to be right. It seems the proportions have been evened out. Likely a small number of large holders moving millions of tokens, mostly into bifrost and scandinodes it seems.

I noticed A-FTSO's VP was down around 30m, I'm sure they'll pick a lot of it back up in them coming days.

It will be interesting to see where the votes go as more exchanges distribute the tokens.

So far there's 4 more Signal Providers added this week.

Edited October 12, 2021 by FTSO_AU

[BillyOckham]

1 hour ago, FTSO_AU said:

That's the big question. It all comes down to being in the median price range, they've managed to do it more than us.

Thanks.  Can I ask how much is the variance generally?  (It’s just pure interest….  I don’t need to know but wonder how tight this all is)

 

Like if the median price is a dollar,  is the cutoff $1.23 or is it more like $1.00023 ?   Or another way of asking is what is the percentage difference between the rewards cutoff figure and the median price?

I apologise if it’s just the dorky folk who are interested in this aspect….   I know most probably won’t care about that. 

[Seoulite]

3 hours ago, FTSO_AU said:

I noticed A-FTSO's VP was down around 30m, I'm sure they'll pick a lot of it back up in them coming days.

It will be interesting to see where the votes go as more exchanges distribute the tokens.

So far there's 4 more Signal Providers added this week.

I reckon this is because as Billy said, bigger holders have much more to lose from diluting their rewards, and furthermore they can take unilateral actions to move below the threshold, whereas for smaller holders like us the difference is negligible and we can't do much about it on our own. 

[ftso_au]

On 10/12/2021 at 1:30 PM, BillyOckham said:

Thanks.  Can I ask how much is the variance generally?  (It’s just pure interest….  I don’t need to know but wonder how tight this all is)

It obviously differs depending on the token in question, but we're talking fractions of a cent in some cases, can be the difference between being in the median or out. There's still a lot to learn and we're doing it in live battle conditions.

I keep coming back to Bifrost as the example, they've got 200m votes so not lacking VP, but they're sliding down the rankings fast. What's changed to cause that? A week ago they were a lock with Alex and Scandinodes, not the case this week.

Edited October 13, 2021 by FTSO_AU