Jason James said:If two players end up tied, the tiebreaker goes to whoever had the lower maximum point score in their predictions..
Basically, if my highest prediction for any single driver hits 15 points, but stormybadger8 only went up to 12, then his picks are actually better than mine. It means his predictions were more consistent and closer to the actual results.
stormybadger8 said:Just give us a casual breakdown and feel free to copy the scoring rules if you think they need to be the same.
And if you don't mind, run the numbers for that Barcelona race so we can see how it all looks in practice.🙂 They might seem similar at first glance, but just because they look alike doesn't mean one is better than the other. Honestly, they might not even be as close as they seem.
They are similar. However, I believe my proposal is superior because it captures forecasts with much greater precision. His suggestion relies on certain rounding methods that simply aren't necessary.
The only adjustment I would consider in my model is perhaps capping the points awarded for misses at 5, 6, or 7. Why? Because if you pick a driver who crashes out in the very first lap to win, you lose 24 points instantly. You're left with zero chance of recovery, regardless of how accurate your overall forecast was. Or what happens if a driver fails to even start? Or some other outlier scenario...
In the current calculation, almost everyone would end up with 15 to 20 points for picking Jenson Button, making them look worse than someone who had a much weaker forecast but correctly predicted a DNF for him.
EDIT: It’s essentially the same principle as statistics: you have to filter out the extreme outliers.
William Bennett53 said:I didn't see much point in drafting an entire set of rules, seeing as they would likely remain identical with the exception of how points are tallied. However, I can offer this one single clarification, if only to ensure everyone understands the scoring logic.
Furthermore, if you are suggesting that Matthew Young35's criteria and my own proposal for the rules are remarkably similar—and given that Matthew Young35 developed the software specifically to aggregate data according to his own criteria—then I find myself quite inclined to support his suggestion. It facilitates a more equitable and logical scoring system while allowing for rapid, seamless automation. 🙏 👍
Forget the software. I could easily write a program for your version too; for a few hundred bucks, anything is possible 😁.
In fact, when I really think about it, the hardest part would be coding for our current scoring method, which happens to be the most imprecise one we have.
Perhaps it is time we actually sit down and have a serious discussion about overhaulng the entire scoring system. We need strict, unambiguous rules that leave no room for manipulation or conflicting interpretations.
stormybadger8 said:Not me either.🤷 But it just makes sense for you to be the one proposing them.😉
If someone actually has an idea, they can easily type it out for you.🙂 Right now, my brain is a total blank.🤷 That previous prediction I made was honestly pretty stupid; we need something smarter than that, but currently, I’ve got nothing.
Maybe we could use exact positions, like we did this time.🤷 Or maybe just whoever has the smallest gap between their prediction and the actual results.
7. In the event of a tie in points, we look at the first 15 predicted positions, then the top 10, and finally the top five. If everything is still tied after that, then all bets are off.
Alright, here is a preliminary draft of the proposed rules.
1. To be considered valid, a prediction cannot list the same driver in multiple positions. Additionally, your entry must include exactly the number of drivers specified by the FIA as participating in the race. 2. In terms of ranking, players with fewer total points will be placed higher than those with more points. 3. Scoring is calculated based on the absolute difference between a driver's predicted position and their actual finishing position. This applies to all selected drivers. 4. If a driver you predicted fails to make the grid entirely, you will be awarded 24 points (representing the maximum possible number of participants). 5. The final standings from 1st to 24th will be determined by the official FIA results, which include DNFs, DNQs, and any other technical disqualifications. 6. PLEASE NOTE: These figures—whether we are looking at 22, 24, or 26 drivers—are subject to change depending on how the FIA settles the regulations regarding the total number of cars on the grid. 7. In the event of a tie in total points, we will break it by looking at the accuracy of the top 15 predicted positions, then the top 10, and finally the top five. If we are still stuck in a deadlock after that? Well, then all bets are off.
stormybadger8 said:I get what the point of the system is, and honestly, I think it’s a "fairer" way to go. But it still needs some fine-tuning. That’s exactly why I want to see how it actually plays out in the real world.😉
And you seriously can't have a system where there isn't a predetermined reason why someone ends up better at equalization.😳 It’s easy enough to end up with a tie where multiple players have the exact same score. We seriously need to decide ahead of time who takes the win and why. If we don't, people start arguing, things get heated, and the whole thing just turns into total chaos. 🙂
That’s exactly my point—we really need to flesh this part out and cover all the bases. For instance, if we look at something like: Someone went and wrote the same driver in two different spots. Someone forgot to list a driver. Someone listed way too many drivers. Looks like someone’s tied on points. and so on.
Look, just try to actually think this through and put some ground rules together so we finally know where we stand. And if anyone’s got an idea, just throw it out there. Feel free.🤷
I mean, it’s pretty obvious that you’re the one who gets to decide the rules since you’re the one who made them up in the first place.😉
Someone went and listed the same driver in two different spots. Someone forgot to mention a driver. Someone went and listed way too many drivers. I’ve been giving this some thought, and I think it might be time to put together some formal rule suggestions.
Does anyone else notice that we have a tie in the points standings? For a situation like this, we really need to come up with an entirely different set of rules and a completely new scoring system. As for that? I’ve already said my piece—I'll leave it to someone else to figure out.
I honestly can't see a single reason why everyone shouldn't be involved in proposing rule changes. Why leave it to a select few?
stormybadger8 said:That’s just not on the table.😳 We definitely won't see any changes to the scoring system this season.☕ The only way it happens is if there's a total consensus—meaning a literal 100% vote from every single player in the league. If even one person is against it or just stays silent, then it's a non-starter for this season. Which, let's be real, means it's basically dead for this year.☕
Any potential changes would only apply to next season, and even then, only if they actually work and we get a massive majority—say, at least 75%—voting in favor.
I couldn't agree more. It would be nothing short of absurd to overhaul the regulations in the middle of a season—unless, of course, every single player on the field was demanding the change.
And one more thing. Throughout this entire discussion, the primary objection to my proposed scoring system was always how difficult it would be to calculate on the fly. That’s no longer an issue. With the pace method, the math is actually dead simple.
It’s actually quite straightforward. My system focuses on the gap between where you predicted a driver would finish and where they actually landed. Essentially, the higher the discrepancy, the fewer points you earn. As for DNFs, we just go by the official FIA standings.
That said, the system could use a little fine-tuning. For instance, I’d cap the maximum score at 5 points per driver. We don't want one poorly placed pick—like someone predicting a driver who ends up retiring from the race—to completely tank a person's entire forecast.
If there's a tie in points? Honestly, I haven't a clue; I'll leave that bit to someone else to figure out.
I'm pushing for this specific scoring method because it feels the most equitable. In my view, it provides the most accurate reflection of who truly has the better intuition. Consider this example: if you predict a driver will take 4th and they finish 3rd, under the current rules, you get nothing. Under my system, you'd walk away with 1 point. Meanwhile, if someone else predicts that same driver for 1st, they missed by two spots, yet under the current setup, they're awarded 5 points. By my logic, the person who misses by a single position should be rewarded more than the one who misses by two.
So, if I'm following you correctly, you can run those numbers super fast using that system.🙂 But how did you handle the issue where everyone spells the drivers' names differently? Like, do you have to manually type out every single prediction from the forums, or does the software just let you copy-paste everything in?
"1. Enter the actual standings into the left table. 2. Use the mouse to drag drivers from the left table over to the right table until the right table reflects the players' rankings. 3. Hit "CALCULATE," and the result pops up in text edit, automatically saving to your clipboard. 4. Paste that clipboard content straight onto the forum. 5. Press the "CLEAR" button to wipe the right table (the left table with the real standings stays put, so no need to re-type it). 6. Repeat step 2 for every remaining player."
In short: you write the real standings once, and after that, it's all just "drag and drop."
1. Enter the actual standings into the left-hand table. 2. Drag the drivers from that left table over to the right one to reflect the player rankings. 3. Hit "CALCULATE." The results will pop up in a text editor and automatically copy themselves to your clipboard. 4. Paste that clipboard content directly onto the forum. 5. Click the "CLEAR" button to wipe the right table clean (the left table with the official standings stays put, so there's no need to re-enter everything). 6. Repeat step two until you've reached the last player.
Once the official standings are entered (about a minute), it takes roughly one minute per player to process.
Here is a suggestion for refining the scoring criteria. I’m proposing that we factor in any successful PP and FL as a -5 point penalty.
EDIT: Actually, I have an even more granular idea involving starting grid positions and fastest lap rankings, but honestly, that would be absolute madness.
Terry Carter2 said:Matthew Young35 mentioned he was going to put together a program to calculate that specific criterion! 😁
ruggedfox28 said:Yeah, well, he said that last year too, and nothing ever came of it... 😬
..and even if he did whip up some little script for that rule, what then? Does every future commissioner need to download his software just to be eligible to run this thing? Because it feels like that rule gets invoked constantly?!
..let's get real here and simplify how we run this competition. It's getting way too complicated and bloated. Honestly, if it wasn't for you, Terry Carter2, doing most of the heavy lifting, this whole thing would have wrapped up in three days...
The spreadsheets are massive, and there's room to screw up at every single turn. We need to automate as much as possible. Without these automated Excel sheets, this would easily drag on for over a week🤣 🤣 ☕
Well, I actually finished about 99% of the program, but it felt like nobody was interested in using it. And yes, obviously they would have to download it... how else could they use it?
