Horses - the new not-really-betting betting game
58 watchers
Nov 2024
12:05pm, 25 Nov 2024
31,285 posts
|
fetcheveryone
Sorry for all the inevitable cross Through no fault nor subterfuge of his own doing, 12 of the top 20 biggest wins were bets placed on @57.5 Degrees of Pain - including the monster 13.5M win. I hope he won't mind me using his stats as an example. In the last 90 days, this lovely thoroughbred had run 10k three times, with a best time of 84:XX minutes. On the weekend, he ran 56:XX, netting a shedload of cheese for his backers. It's some way off his impressive best ever of 36:XX, but we have to go back 15 months to find a quicker time for him. So on the one hand, it's really lovely that (perhaps) the game has encouraged a runner to go out and give it some welly. And on the other hand, we need to find a way to filter out horses where the 90 day benchmark is not reasonably indicative of their abilities. Illness or injury can easily eat up a big chunk of 90 days, and leave some very low-hanging fruit when the horse returns to fitness. We could either (a) increase the 90 days to [say] 12 months, or (b) make the horse un-bettable unless they'd run the distance a certain number of times. In the case of this horse, if we look back over 12 months, the benchmark would have been 68:XX, which is better but still produces a large return. Alternatively, if we set a minimum of 10 x 10k in the 90 days in order for the horse to be bettable, no bets would have been placed, as the horse has only recorded three such performances. A detractor might say that this might remove the incentive for the horse to have run the quicker 10k. But if we say that the thresold for 5k is 15 performances, the horse achieves this comfortably: 38 5k+ runs during the 90 days, with a benchmark of 22:XX, which seems fairly reasonable. So the horse would still be able to attract bets at 5k, and could be incentivised to 'collect' enough 10k performances to be considered for betting. Thoughts? |
|
Nov 2024
12:05pm, 25 Nov 2024
31,286 posts
|
fetcheveryone
*cross posts 
|
Nov 2024
12:15pm, 25 Nov 2024
25,417 posts
|
57.5 Degrees of Pain
fetcheveryone wrote: I was pretty much in category 1. I hadn't run 10k at one time much this year recovering from illness/injury. But my few longer runs were with friends and involved a few rest stops. The game did provide the incentive to log a full 10k with only the odd walk up the steepest hills, but a huge 'PB' was virtually assured. I also publicised the plan.
I agree that the returns need to be scaled back a bit - but I need to make sure I don't throw the baby out with the bath water. I'm interested in finding out the scenarios that have thus far yielded enormous wins. Here are some situations that I think should be coded out of the game: 1) a person has a 10k benchmark from a single run-walked attempt in (say) 90 minutes, and then goes out and runs 10k in (say) 60 minutes. Massive chunk off, huge win for any gamers that spot it. I could make this far less likely by making sure that the horse has recorded a set number of attempts at this distance over the last 90 days e.g. has covered 10k at least 10 times. Whilst they *could* all have been run-walked, it seems much less likely. This threshold number of performances could be tied to distance e.g. 30 x 1 mile, 15 x 5k, 10 x 10k, 5 x half marathon. 2) a (well-intentioned) player says "I'm about to upload, get your bets in". Maybe the return could be based on the number of hours since the bet, or at least zeroed out if doesn't predate the activity start date. This might take a bit of thinking when it comes to time zones. |
|
Nov 2024
12:16pm, 25 Nov 2024
25,418 posts
|
57.5 Degrees of Pain
Excellent cross post there. Now going to read myself as an example!
|
|
Nov 2024
12:21pm, 25 Nov 2024
31,287 posts
|
fetcheveryone
If we look at another horse, @chunkywizard - he netted a load of cash for his backers by taking 94 seconds off his five mile time at a race on Sunday. The biggest win was a 126345 credit return from a 3000 credit stake. In this case, the horse has done this distance plenty of times (42) in the last 90 days - so the previous benchmark was not unreasonable. And in terms of what the horse is capable of, the last two years show only six quicker times for this distance, in 238 runs. So in this case, I'd say reducing the return by a factor of 100, making it 1264 for a 3000 credit stake, makes it a not unreasonable outcome. |
|
Nov 2024
12:34pm, 25 Nov 2024
31,288 posts
|
fetcheveryone
And looking at some more big wins, there was a return of 50,490 credits on a bet of 1000 credits on @ptr_runner at 5k. He had 59 5k times in his 90 day history, and took ~67 seconds off his benchmark. This seems reasonable given that in the last two years, he's only gone quicker than that on four occasions. Making the return 505 credits on a 1000 credit investment seems like a reasonable amount of risk. |
|
Nov 2024
2:52pm, 25 Nov 2024
31,289 posts
|
fetcheveryone
Right then. I've reset all punters back to 1000 credits, and removed all bets. If you visit the game as a punter or a horse, you can claim your 20/50 visiting bonus for the day, even if you've already done it once today. The return on bets has been reduced by a factor of 100; and I've added some minimum thresholds for the number of performances you need to accrue at any given distance. For example, you need to have run twenty 5k+ runs in the 90 day period in order to be bettable on as a 5k horse. This tapers down, so that you need ten 10k runs; five 10 mile runs; three HM distance runs; or one prior marathon to appear as a bettable horse in the given distance category. |
|
Nov 2024
4:01pm, 25 Nov 2024
25,423 posts
|
57.5 Degrees of Pain
fetcheveryone wrote: That all sounds sensible.
Right then. I've reset all punters back to 1000 credits, and removed all bets. If you visit the game as a punter or a horse, you can claim your 20/50 visiting bonus for the day, even if you've already done it once today. The return on bets has been reduced by a factor of 100; and I've added some minimum thresholds for the number of performances you need to accrue at any given distance. For example, you need to have run twenty 5k+ runs in the 90 day period in order to be bettable on as a 5k horse. This tapers down, so that you need ten 10k runs; five 10 mile runs; three HM distance runs; or one prior marathon to appear as a bettable horse in the given distance category. |
Nov 2024
4:32pm, 25 Nov 2024
9,706 posts
|
Pothunter
Is 20 times 5k in 90 days in order to be able to bet a bit steep? If you do parkrun every week then that would be 12 attempts (I think!). I guess most people doing parkrun might also train occasionally (shock horror!) but fair to assume that those runs would almost certainly be slower than the parkrun attempts so wouldn’t likely influence the best time.
|
|
Nov 2024
4:37pm, 25 Nov 2024
31,290 posts
|
fetcheveryone
69 of the 105 horses currently have enough times; and a further 12 have between 15 and 19 runs (inclusive).
|
Useful Links
FE accepts no responsibility for external links. Or anything, really.Related Threads
-
Trader: New Game
Mar 2025
-
Unlockable Badges
Mar 2025
-
Fetch Treasure Hunt
Mar 2025
-
New Fetch Game: Hide and Seek
Feb 2025
-
Fetch Virtual Challenge 2025
Feb 2025
-
Who Squares Wins!
Feb 2025
-
FetchPoint: The Game
Feb 2025
-
Fetch Virtual Challenge 2024
Feb 2025
-
Rundle
Feb 2025
-
Who Squares Wins (WSW): Records and Trivia
Feb 2025