|
Post by vbbetterthanbb on Dec 8, 2014 19:18:40 GMT -5
Since a few people are interested, scaled scores will be provided after our official PTW-NCAA winner has been crowned. I actually believe the one point per correct pick is the fairest scoring method. The scaled scoring by rounds of 1-2-4-8-16-32 puts prohibitive emphasis on correctly picking the eventual NCAA champion, leaving those who have not picked the champion virtually no chance of winning the bracket contest. In that sense a scaled scoring such as 1-2-3-4-5-6 would be better. Even then, it still compounds weighted penalty against incorrect picks in the early rounds. Which to me, eliminates too many participants too early in a bracket contest -- no fun. Well, that's just my opinion. You are welcome to express yours .
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 8, 2014 22:52:05 GMT -5
My bracket game is weak. Should have chosen based on mascot.
|
|
|
Post by psuvbfan10 on Dec 8, 2014 23:55:28 GMT -5
Choose bracket based on best dressed coach?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 9, 2014 11:02:22 GMT -5
This year the dog chose.
|
|
|
Post by smb4 on Dec 9, 2014 11:54:37 GMT -5
Since a few people are interested, scaled scores will be provided after our official PTW-NCAA winner has been crowned. I actually believe the one point per correct pick is the fairest scoring method. The scaled scoring by rounds of 1-2-4-8-16-32 puts prohibitive emphasis on correctly picking the eventual NCAA champion, leaving those who have not picked the champion virtually no chance of winning the bracket contest. In that sense a scaled scoring such as 1-2-3-4-5-6 would be better. Even then, it still compounds weighted penalty against incorrect picks in the early rounds. Which to me, eliminates too many participants too early in a bracket contest -- no fun. Well, that's just my opinion. You are welcome to express yours . I did a bball bracket years ago in which I gave extra points for correctly picking underdogs (I don't remember the formula). This didn't resolve the problems you've identified, but it did encourage people to take more chances. Just throwing that out there - I am NOT asking you to do it. I appreciate the time you already spend on this.
|
|
|
Post by vbman100 on Dec 9, 2014 12:27:07 GMT -5
I like the scaled scoring better. I think the 1-2-3-4-5-6 is ok but somewhere in between that and the double points (1-2-4-8...) each round is best. I am not a fan of the 1 point per match scoring (unless I win this thing). That weighs the 'Penn St vs cupcakes' picks the same as picking the correct winner of 'PSU vs Wisconsin' or 'Stanford vs Washington'.
Also, if I am 5 or 6 points behind now, there are only 15 matches left, and the chance of me winning with a 1-1-1... system is slim. With scaled scoring, if I have some variation with just 2 or 3 picks or 1 FF pick, I have a shot.
|
|
|
Post by c4ndlelight on Dec 9, 2014 12:37:55 GMT -5
Or, we could do it the Big XII way and just declare two champions EVERYONE GETS A TROPHY
|
|
|
Post by vbbetterthanbb on Dec 9, 2014 16:24:28 GMT -5
Choose bracket based on best dressed coach? Bye Rose, the blue sweater won't cut it. Not lion? I did a bball bracket years ago in which I gave extra points for correctly picking underdogs (I don't remember the formula). This didn't resolve the problems you've identified, but it did encourage people to take more chances. Just throwing that out there - I am NOT asking you to do it. I appreciate the time you already spend on this. Perhaps a scaled score based on the % of participants' pick for each match? E.G. those who pick UCLA in "28. Penn State v. UCLA" (PTW picks are 86 v. 5) would get more points if UCLA wins than those who pick PSU if PSU wins? However, that does get complicated, and then the point assignment is still arbitrary. ...Also, if I am 5 or 6 points behind now, there are only 15 matches left, and the chance of me winning with a 1-1-1... system is slim. With scaled scoring, if I have some variation with just 2 or 3 picks or 1 FF pick, I have a shot. Guess what, if all your actual remaining picks work out, you would be the winner in the 1-1-1-1-1-1 format, but NOT in the 1-2-4-8-16-32 format! I didn't simulate your score in the 1-2-3-4-5-6 format, but would bet that you lose in that one too. Or, we could do it the Big XII way and just declare two champions EVERYONE GETS A TROPHY We are on limited budget -- NOBODY GETS A TROPHY . The official winner though, gets to proudly travel around the vast VT kingdom with his/her imaginary scepter for the next 12 months ! Hmmm... with Texas hogging all the cash, how was Big XII able to afford 2 trophies?
|
|
|
Post by vbbetterthanbb on Dec 9, 2014 18:17:48 GMT -5
I like the weighted system ...but I wasn't sure if I could say why I liked it. There's pros and cons to either a weighted scoring or a single-point scoring... 1/2/3/5/8/8... 2/3/5/8/12/12... Sorry I did not even see your post until now. I like the single-point system because all the pre-determined weight scoring are just too arbitrary, and therefore could be subject to endless debate after a contest is over. As mentioned above, an alternative model that awards points based on the odds of the actual teams involved in each match seems to make more sense, though it would still involve some arbitrary (but IMO much less so) point assignment. ...if I have some variation with just 2 or 3 picks or 1 FF pick, I have a shot. I was right, if all your remaining picks were correct, you'd also lose with the 1-2-3-4-5-6 system. I would further venture to guess that I would have to try really hard to come up with a weighted scoring system where your current picks would come out as a winner! So, what bribe can you offer ?
|
|
|
Post by vbman100 on Dec 10, 2014 9:02:05 GMT -5
...if I have some variation with just 2 or 3 picks or 1 FF pick, I have a shot. I was right, if all your remaining picks were correct, you'd also lose with the 1-2-3-4-5-6 system. I would further venture to guess that I would have to try really hard to come up with a weighted scoring system where your current picks would come out as a winner! So, what bribe can you offer ? Well, you have convinced me to change my mind. I love the 1-1-1.. system way better than any stupid weighted system! Now that I think I have a chance to win, I am going to be nervous all weekend. So much pressure. Thanks. You're really good at separating quotes in replies. You'll have to teach me how to do that someday.
|
|
|
Post by smb4 on Dec 10, 2014 16:54:19 GMT -5
I did a bball bracket years ago in which I gave extra points for correctly picking underdogs (I don't remember the formula). This didn't resolve the problems you've identified, but it did encourage people to take more chances. Just throwing that out there - I am NOT asking you to do it. I appreciate the time you already spend on this. Perhaps a scaled score based on the % of participants' pick for each match? E.G. those who pick UCLA in "28. Penn State v. UCLA" (PTW picks are 86 v. 5) would get more points if UCLA wins than those who pick PSU if PSU wins? However, that does get complicated, and then the point assignment is still arbitrary. It could be simplified - just give 2 points for an underdog win and 1 point otherwise. Underdogs could be determined by [choose your favorite ranking system] with a range in which the match is considered even. (For example, if you use rpi and the two teams are within 3 of each other in rpi, neither team would be considered an underdog.) This rewards risk-takers and gives them a little buffer when they get it right. Maybe next year...
|
|
|
Post by vbbetterthanbb on Dec 10, 2014 17:21:05 GMT -5
...separating quotes in replies. You'll have to teach me how to do that someday. My steps: 1. Open a word processing (or text editing) software 2. Click on "Quote" for the VT post that's being replied to 3. Copy and paste the content of the Create Post box into the word processing file window 4. Exit the Create Post box by clicking the "go back" arrow key of the browser 5a. Leave the (R-square bracket)quote author=...(L-square bracket) --- which signifies the beginning of the quote b. Edit the post message that's being replied to and leave whatever part desired c. Leave the (R-square bracket)/quote(L-square bracket) --- which signifies the end of the quote 6. Now add the reply to the word processing file 7. Repeat steps 1-4 for additional post(s) to be replied to, and add Create Box content for each of those to the bottom of the the word processing file, then edit each as needed 8. When all done adding/editing, click on Reply at the bottom of the VT thread, copy content of the word processing file and paste into the VT Create Post window 9. Click Preview to make sure the reply looks okay before clicking on Create Post to finish Hope this is not too confusing. I had to do this a few times before I have good control of the results.
|
|
|
Post by vbbetterthanbb on Dec 10, 2014 17:52:53 GMT -5
It could be simplified - just give 2 points for an underdog win and 1 point otherwise. Underdogs could be determined by [choose your favorite ranking system] with a range in which the match is considered even. (For example, if you use rpi and the two teams are within 3 of each other in rpi, neither team would be considered an underdog.) This rewards risk-takers and gives them a little buffer when they get it right. Maybe next year... Sounds reasonable, but that adds multiple complications of incorporating the data from another ranking system, with programming likely the easiest part . At this moment, I am not at all convinced that the 1 point per pick scoring system is not the best, or at least as good as any other scoring system. One previous year I added a side contest to the PTW weekly contest, by allowing participants to put percentage of their picks (e.g. 51% on Stanford and 49% on Oregon State -- okay here comes the hate mails -- calm down, it's just an example). The bottom line, the side contest's results were so very close to the regular PTW contest that it was simply a failed attempt. Anyway, I'll post results of this year's contest in various scoring scales after our official winner is crowned. We'll see then if there's even a compelling need to make any changes at all.
|
|
|
Post by vbman100 on Dec 10, 2014 20:40:29 GMT -5
I thought there was a list of current leaders and their point totals somewhere in this thread. I can't find it. Was it posted somewhere? Am I dreaming? Did you remove that?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 10, 2014 21:03:20 GMT -5
I thought there was a list of current leaders and their point totals somewhere in this thread. I can't find it. Was it posted somewhere? Am I dreaming? Did you remove that? Being the contest administrator, I removed it as I considered it "a spoiler" which would also make the last week of the contest anticlimactic.
|
|