It would be cool if we could see everyones payout on the results page. That way we don't have to click each person to see what they got.
| ioi said: rendo, I was considering that this morning and I think it will work better in cases where the average accuracy is lower so yes that is probably how I will modify the scoring and maybe have a 1.1 in there as well to basically reward people for entering: Winnings = 1.1 * (1 + (your accuracy-average accuracy)/100) * bet This way, you can earn between 0 and 2.2 times your original bet and you only need to be better than 9% below the average accuracy for a particular prediction to break even. |
Sounds good... that's what i call a more balanced formula indeed...
While working this afternoon, another suggestion came to my mind: give a little bonus to the ones who have the best global accuracy...
For an example, in my case, i've bet $200, with a 65% accuracy... so i thought, why not add to your first formula a small bonus of (total amount of bets * total accuracy) / 2... here, it would be (200 * 0.65) / 2 = 65...
The good thing about that bonus is that it would incite people to do predictions on ALL games and consoles, and not do a single bet on one console or one game, and nothing at all for all the rest... if you try for the 10 games and the 5 consoles, and were globally correct, you are rewarded... what do you think about this, ioi?
But otherwise, your new formula is fair enough... even if i think most people won't bet very high on consoles sales in the future, because they'll be sure to loose or not win that much on these...
"A beautiful drawing in 480i will stay beautiful forever...
and an ugly drawing in 1080p will stay ugly forever..."
If you're rewarding global accuracy compared to everyone else, it should be the same as the current system or the one I proposed above with the differences so you only get a slight haul. That'd encourage betting big and aiming for good accuracy. *shrug*
| Soriku said: I won $156.84 w00t. |
Actually you lost $18.16 since you bet $175. Lol
Still better than me.
Consoles I Own: NES, SNES, N64, GC, GBA, DS, X360, Wii
Or better yet, add a 16th bet that is based on your global accuracy. You bet on how will you'll do compared to everyone else. All that money is pooled into one giant pool and is divided up among the top predictors. Not exactly sure how at the moment, but it'll come to me if i think about. All the money put in to the global accuracy prediction should be all that comes out, so you either win or lose.
Edit
The global accuracy prediction is an optional bet.
Yeah if it's possible it would be nice to see each persons payout on the list... as well as the breakdown of payout on each persons page so we can see how much we got/lost for each game/console
| 68soul said: As an example with that 96% PS3 AVERAGE accuracy, here are the results of the best 50 players in the League that week for the PS3 prediction: 73.98 / 48.78 / 84.06 / 86.58 / 93.26 / 93.26 / 86.58 / 48.78 / 86.58 / 61.38 73.09 / 98.62 / 95.78 / 99.18 / 86.58 / 75.61 / 99.18 / 85.51 / 70.55 / 99.18 75.61 / 93.26 / 93.26 / 99.18 / 61.38 / 84.06 / 90.73 / 86.58 / 95.78 / 87.84 99.18 / 75.61 / 86.58 / 99.18 / 75.61 / 61.38 / 95.78 / 99.88 / 61.38 / 65.53 99.18 / 10.96 / 73.09 / 83.17 / 46.25 / 86.58 / 99.18 / 99.18 / 88.21 / 89.1 As you may see, yes, 18 of these are in the 90%, and 10 in the 80%... so half of the players were close... but does it mean everybody was very close? Not at all... I don't have the time to write all the 170+ results, but i'm sure you'll find the same tendency: the AVERAGE prediction doesn't mean most people were right, and that it's a really TOO EASY prediction... it just means most people were in the same kind of numbers... In that case, from 20.000 to 60.000... and it's normal, we all know it will not be bellow, or it will not be higher... so hardware predictions will never be rewarding then... Anyway, on these 50 people, only 10 have more than 96%, and win... nearly nothing... so that system drains money out of the game... Now, i understand your point, ioi: you don't want the people to easily cash in on the easiest predictions... but i think that system should be more balanced... a 90% prediction should give people some good earnings anyway... and a 30% prediction should not give people a lot of earnings, just because most of the others were even more wrong... but once again, it's only my opinion, and i'll have fun doin' my predictions, no matter what the system finally is... |
Oh, btw, i've just done the math: the average accuracy of these 50 predictions is 79%... that's really far away from 96%...
I think that's the main problem: look at the global accuracies you find here for all the gamers in the League...
http://www.vgchartz.com/predictions/results.php?game=106
It's the accuracy of the average prediction, and not the average of all the accuracies... just as in my example here, 96% vs 79%...
I hope it's clear, my english is sometimes too limited to express myself as i really would/should...
But ioi, could you confirm: in your formula, you use the 96%, and not the 79% result as the average accuracy, right?
"A beautiful drawing in 480i will stay beautiful forever...
and an ugly drawing in 1080p will stay ugly forever..."
| rendo said: ( ( (100 + (Accuracy-Average Accuracy) ) / 100) * original bet) That's how the payout should work compared to the comparison of accuracies. This way it's capped in that you can't make mega bucks if everyone screws up the average accuracy on something by underpredicting. 85.67%/40.45% = 2.118 X bet = double your money? Flawedz. |
That formula sounds good.
| Soriku said: I won $156.84 w00t. |
Wow, that's cool. I can see you have 2Kvg$. That's cool. Time to make big bets, soriku. 
About Us |
Terms of Use |
Privacy Policy |
Advertise |
Staff |
Contact
Display As Desktop
Display As Mobile
© 2006-2024 VGChartz Ltd. All rights reserved.
