I also find the current
Monthly Statistics confusing/not relevant. According to january:
---------------------------- Monthly Statistics ----------------------------
UTC Month Pool Avg Blocks Expected Mean Diff% MeanTx% CDF[Erl] Luck% PAPPS%
2018 January 31.98PHs 8 6.87 85.86% 132.04% 0.3817 116.47% 152.40%
what was the number of blocks expected for January? 6,87 ---> Looks like wrong, as the rest of the data for January.
This system of calculating only for last block is showing best luck possible calculated just when block is won. In the case of January it shows to be a extreme case. Of course January has been an very unlucky month... but.. it will always show green (good luck month) historically to the user giving "wrong" quick look information.
Last row (current month) in Monthly Statistics should be calculated daily, at my point of view, with a cron job so it doesn't give every day (except the day when block is found) "uncorrect" info for the month, as it is happening now.
Calculating last row daily would probide exactly the on-going stasts and the final stats for the month. That is, in my oppinion more useful for the user than the info linked to the moment when las block was won. In fact, for me knowing what happened in January, those Monthly Statistics, would seem to be wrong, if I would not know that they are linked to the last block won (in this month case skipping the last almost 10 days (1/3 of the month) BAD LUCK)... and well... As said, even knowing the related "trick" these stats for me seem to provide no use for the specific and extreme case of January.
I am not critisising, I am just sharing my point of view, because I would like a change there... to the daily Monthly Statistics's last row update cron task approach. That would make me happy! :-)
First of all, if you want daily luck projections, use one of the many calculators that are available out there. For example, I just used one and it shows that the expected number of blocks for our average pool hash rate for the 31 days in January was 8.9 blocks. We hit 8 blocks so I'm not sure how you can say January was a "very unlucky month".
Second, The Monthly Statistics table is simply to show the actual blocks hit and how much work went into each one compared to what the expected would be at that instant. It's not just simply another projection into the future...you can do that yourself. Since we had 9 days of work at the end of January where no blocks were found, that work will be included in the first block hit in February's block line.