Saturation curve or metrics are displayed incorrectly

[Leo-Raumann]

Hi,

it seems to me that there is a bug in the saturation curve:
https://pooltool.io/pool/e34717fcd0077c52fc9f2dfa64bdc241c10144dd4a90b081359aa0fd/curve

The ROS is displayed as 0.00% for 0.5 mio stake, although the expected chance for one or more block is displayed as 35.49% (metrics):
https://pooltool.io/pool/e34717fcd0077c52fc9f2dfa64bdc241c10144dd4a90b081359aa0fd/metrics
also, the mouse over tool tip for the bars are inconsistent (always showing 0% in the tool tip title)

and finally, I wonder why the block chance is different to the 44% stated here:
https://cexplorer.io/pool/pool1udr30lxsqa799lyl9haxf0wzg8qsz3xaf2gtpqf4n2s064jrvsj/blocks#data

Thanks!

[papacarp]

good morning, thank you for the feedback. Indeed there are some issues with small stake pools. The goal of this chart was to illustrate the saturation curve near the other end of the spectrum, but you are indeed right that it painted the pools with <1M stake as to not have rewards, which is wrong.

1. I have adjusted the calculations to give an approximate expected long term return for pools that make less than 1 block per epoch. but its better than showing zero return. I'll be pushing changes up during our routine maintenance tonight.
2. regarding chances, I'm not sure how they calculate chance. A pool with 0.5M ada is expected to make ~0.429 blocks per epoch right now but the chance of making 1 or more blocks according to our binomial distribution calculations is around 35%.
3. Regarding tooltip titles, i can't replicate that issue. Can you perhaps take a screenshot of the problem or let me know what browser you are using?

[Leo-Raumann]

hi,

1. thanks for your quick reply and the adjusting.

2. Oh sorry, I didn't pay attention to the different units (average blocks vs chance for one or more block). Thanks for pointing out.

3. Regarding the tooltips, I made some snippets via [windows key + [shift] + s

First it looks like this:

after 2 seconds the page reloads and then it looks like this:

tested browsers:
windows 10 pc: modzilla firefox, chrome
mobile: mi browser

Cheers
Leo

[papacarp]

strange. just tried with firefox v105 and v100 on both windows 11 and windows 10 and can't see the 0% chance of 1 blocks thing.

The 2 second delay is just how the page loads. Takes a few seconds to run the simulations and chart everything.

I'll keep looking for this as I get time today.