Comment by 🕹️ skyjake

I think that should make the tournament a little more even, which is good.

Are you opposed to the top four getting points? With 9-10 daily players, would be good to give out points to more than three of them. But instead of giving out 4 points to the top player, maybe keep the spread still between 1..3 (like 3, 2, 2, 1).

2024-06-21 · 1 year ago

10 Later Comments ↓

hmm interesting proposition. accounting for ties in this situation works be the hardest part. what about 3,2,1,1?

I honestly don't understand the value of making things more even. Is it to make it easier to catch up? Because I'm not sure that it would. What it would do is value someone who consistently makes third place over someone who makes it to the top every third time; and that's not clearly a good thing to me.

I guess if any changes are made it would be good to be clear about what they are supposed to acheive and why that's good.

yep, that's why I put it out here, to get some inputs. the intent was to reduce potential run away monthly matches.

but talking about run away matches, it could be that that person just played more days in the month.

@gritty I was thinking the same wrt. playing every day vs. less frequently. To do well in the tournament, you basically have to keep playing every day. In effect, someone who does well in the tournament therefore is 1) a regular daily player, 2) has a few lucky breaks, and 3) balances risk optimally.

Given that actual skill is only involved with 3), I wonder how the scoring could highlight that aspect, or in other words, mitigate the effects of 1) and 2).

One option might be to simply give 1 point to the top 50% of players, but I suck at probability math and statistics, so I'm not sure how that would actually affects things. 🙂

In any case, I think it's a good idea to try a few different scoring schemes and see what seems to work the best.

@gritty I tied for first today

It’s a good opportunity to see if the tournament logic is working as you expect.

@skyjake I like the idea of trying a couple different scoring systems and seeing which works best. Also maybe later today I will do some thinking about different systems and their advantages and disadvantages and write a gemlog post.

Personally, I think that playing every day is something that the tournament system should encourage.

One important question to me is this: should someone who performs consistently well but never exceptionally (someone who takes few risks in other words) win the tournament over someone who risks losing and occasionally wins big.

I guess I feel that both strategies should be viable.

What about gaining points same or similar to how it is now on a per game basis but use the players mean score for the tournament? The better and more consistently you place then the higher you'll be in the rank. But frequency of play over the month wouldn't incur any significant advantage.

@satch saw your post on DSN. thank you

@satch I had an idea similar to your dynamic point allocation.

assign points based on your percentage of the sum of everyone's score. If 20 points are up for grabs, first place would have the largest percentage, rounded to the nearest whole number.

I ended up not messing with the points. One, I ran out of time, and two, I want to see how it goes now that the average number of regular players has increased a bit.

Original Post

I'm considering changing the points to 3,2,1 instead of 5,3,1 for the tournament scoring, starting next month. thoughts?