Live Chat With Greg A. On Wednesday, 11 a.m. EST

[BillyWaz]

Originally posted by Gordon Gekko:
quote:Originally posted by BillyWaz:
For the record Greg, I think it is a GREAT idea to offer a BIG prize for someone who wins both. If both contests have 400 people, the odds are 1:160,000. I think an insurance company would see that, and it wouldn't be too much to purchase, but I am just a Special Ed Teacher and just guessing.
Waz - u musta failed stats in school. The odds of the insurance company paying out the million would be WAY lower than that. WAY LOWER [/QUOTE]I was ASSUMING that all 400 would play in both (which is all but unlikely I admit), and I simply DID do the fact that we are all "ping pong balls". A lot going on, and I was using theoretical probability.

I didn't take a stats class (Special Ed), but you may be right that I probably would have failed.

[ May 15, 2008, 03:50 PM: Message edited by: BillyWaz ]

[kjduke]

The approach to this from an odds/actuarial viewpoint would be looking at Player X (who will be the winner of one league) and his/her probability of winning the second contest.

Considering co-variances of player X's luck and skill in winning the first contest, i.e., the chances of his/her best performers overlapping and his/her skill relative to other players, I would say it fair to call Player X's chances of winning a second similar contest as being equal to that of at least the top quartile of players.

So if there are 400 players in each contest, the odds of player X (the player who will win one contest) winning the second contest would be 100:1. A very conservative view (from an insurance standpoint) could put that player in the top decile, thus reducing the expected odds to 40:1.

The two scenarios would yield an expected payout of either $10,000 (scenario 1) or $25,000 (scenario 2). Given the lack of adequate history, I'd expect the conserative view to be closer to what an actuary would price in.

Also, assuming this specialty risk likely would result in a significant profit margin requirement (I'd assume a 50% expected profit margin), my best guess of insuring this million dollar payout would be $50,000

... continuing on, from the NFFC/NBC business standpoint, this would cost $62.50 per entrant, or just under 1/5th of their incremental margin per entrant. They'd need to attract appx 150 additional entrants to breakeven on the proposition. Of course, if you consider an entrant an annuity with a 3-yr avg lifecycle, then you'd need to attract only 50 new entrants for a 3yr breakeven. Not to mention the early-mover advantage and marketing edge in growing the business bigger, sooner.

All in all, I'd say it's well worth the NFFC/NFB to offer at this cost level.

[ May 15, 2008, 06:01 PM: Message edited by: KJ Duke ]

[RiFF]

Kevin, you assume 100% overlap between the 2 contests which necessitates the winner from one contest being entered into the sister contest at 40-1 odds.
I have it from a reliable source that the projected winner of the NFFC 2008 contest will not be entered into the 2008 NBC contest. Therefore, the underwriter at the specialty ins. co. need not lose any sleep. By the way, at a $50,000. premium level I'd like to be part of that ins. consortium.
Its probably fair to say some percentage of contestants have less than an equal chance of winning; but at this level, I'd say most of the remainder are pretty evenly matched. That's to say, for instance, I don't believe any of the former winners of this contest would have a 40-1 chance of winning this year. Assuming 396 entrants I would say any one of their chances are much better than 396-1, but not close to 40-1 (or 100-1).

[Baker Boy]

Excellent analysis KJ, I would tweak the premium down a bit (from 50k) based on the % of players playing both events. I believe that is something Riff touched on earlier. The 50k should be based on 100% duel play.

Of course if someone were to actually win this this year to take full advantage of the marketing it would be have to be offered next year. The premium could be 250k. Gregg dont't worry you'll have a thousand players in both leagues to cover it.

[ May 15, 2008, 06:05 PM: Message edited by: Baker Boy ]

[pizzatyme]

Originally posted by KJ Duke:
The approach to this from an odds/actuarial viewpoint would be looking at Player X (who will be the winner of one league) and his/her probability of winning the second contest.

Considering co-variances of player X's luck and skill in winning the first contest, i.e., the chances of his/her best performers overlapping and his/her skill relative to other players, I would say it fair to call Player X's chances of winning a second similar contest as being equal to that of at least the top quartile of players.

So if there are 400 players in each contest, the odds of player X (the player who will win one contest) winning the second contest would be 100:1. A very conservative view (from an insurance standpoint) could put that player in the top decile, thus reducing the expected odds to 40:1.

The two scenarios would yield an expected payout of either $10,000 (scenario 1) or $25,000 (scenario 2). Given the lack of adequate history, I'd expect the conserative view to be closer to what an actuary would price in.

Also, assuming this specialty risk likely would result in a significant profit margin requirement (I'd assume a 50% expected profit margin), my best guess of insuring this million dollar payout would be $50,000

... continuing on, from the NFFC/NBC business standpoint, this would cost $62.50 per entrant, or just under 1/5th of their incremental margin per entrant. They'd need to attract appx 150 additional entrants to breakeven on the proposition. Of course, if you consider an entrant an annuity with a 3-yr avg lifecycle, then you'd need to attract only 50 new entrants for a 3yr breakeven. Not to mention the early-mover advantage and marketing edge in growing the business bigger, sooner.

All in all, I'd say it's well worth the NFFC/NFB to offer at this cost level. Yeah, what he said!

[kjduke]

Originally posted by RiFF:
Kevin, you assume 100% overlap between the 2 contests which necessitates the winner from one contest being entered into the sister contest at 40-1 odds.
I have it from a reliable source that the projected winner of the NFFC 2008 contest will not be entered into the 2008 NBC contest. Therefore, the underwriter at the specialty ins. co. need not lose any sleep. By the way, at a $50,000. premium level I'd like to be part of that ins. consortium.
Its probably fair to say some percentage of contestants have less than an equal chance of winning; but at this level, I'd say most of the remainder are pretty evenly matched. That's to say, for instance, I don't believe any of the former winners of this contest would have a 40-1 chance of winning this year. Assuming 396 entrants I would say any one of their chances are much better than 396-1, but not close to 40-1 (or 100-1). Assuming an 80% chance that Player X is entered into both contests would reduce odds to 50x and 125x, respectively. That's reasonable. The better scenario would be for one of GE's insurance subsidiaries to underwrite the risk at cost, then we're talking a range of just $8,000-$20,000.

[ultimatefs]

Originally posted by KJ Duke:
quote:Originally posted by RiFF:
Kevin, you assume 100% overlap between the 2 contests which necessitates the winner from one contest being entered into the sister contest at 40-1 odds.
I have it from a reliable source that the projected winner of the NFFC 2008 contest will not be entered into the 2008 NBC contest. Therefore, the underwriter at the specialty ins. co. need not lose any sleep. By the way, at a $50,000. premium level I'd like to be part of that ins. consortium.
Its probably fair to say some percentage of contestants have less than an equal chance of winning; but at this level, I'd say most of the remainder are pretty evenly matched. That's to say, for instance, I don't believe any of the former winners of this contest would have a 40-1 chance of winning this year. Assuming 396 entrants I would say any one of their chances are much better than 396-1, but not close to 40-1 (or 100-1). Assuming an 80% chance that Player X is entered into both contests would reduce odds to 50x and 125x, respectively. That's reasonable. The better scenario would be for one of GE's insurance subsidiaries to underwrite the risk at cost, then we're talking a range of just $8,000-$20,000. [/QUOTE]I doubt 50% would enter both contests. I'll be surprised if it's over $12K.

One could easily provide the data from four NFFC years and 6-7 WCOFF years and look at where the winners of one year have finished all the other years and would determine the odds are remote.

No one could possibly draft anything close to the same team in each event as there are a different number of teams and each draft is so different. Even more so this coming year.

I hope it gets done. It would be an incredible marketing tool.

[ May 15, 2008, 06:36 PM: Message edited by: JohnZ ]

[RiFF]

Originally posted by KJ Duke:
quote:Originally posted by RiFF:
Kevin, you assume 100% overlap between the 2 contests which necessitates the winner from one contest being entered into the sister contest at 40-1 odds.
I have it from a reliable source that the projected winner of the NFFC 2008 contest will not be entered into the 2008 NBC contest. Therefore, the underwriter at the specialty ins. co. need not lose any sleep. By the way, at a $50,000. premium level I'd like to be part of that ins. consortium.
Its probably fair to say some percentage of contestants have less than an equal chance of winning; but at this level, I'd say most of the remainder are pretty evenly matched. That's to say, for instance, I don't believe any of the former winners of this contest would have a 40-1 chance of winning this year. Assuming 396 entrants I would say any one of their chances are much better than 396-1, but not close to 40-1 (or 100-1). Assuming an 80% chance that Player X is entered into both contests would reduce odds to 50x and 125x, respectively. That's reasonable. The better scenario would be for one of GE's insurance subsidiaries to underwrite the risk at cost, then we're talking a range of just $8,000-$20,000. [/QUOTE]This is obviously all speculation, but I'd be really surprised if there was much more than a 50% cross-over from NFFC to NBC. Greg would certainly like 80% but I just don't see that kind of number. My concern, if I was greg, would be having the NBC contest cannibilizing contestants from the NFFC contest. If that happened, the percentage of overlap might increase but at the expense of the number of entrants into NFFC.
And I still don't see the odds of the winner of contest A being much better of winning contest B than the odds of the other entrants in contest B; excluding the 20-25% of "dead-money" participants.
And I agree, the most efficient way to handle this is probably through an GE sub.

[mkrucek]

Originally posted by JohnZ:
quote:Originally posted by KJ Duke:
quote:Originally posted by RiFF:
Kevin, you assume 100% overlap between the 2 contests which necessitates the winner from one contest being entered into the sister contest at 40-1 odds.
I have it from a reliable source that the projected winner of the NFFC 2008 contest will not be entered into the 2008 NBC contest. Therefore, the underwriter at the specialty ins. co. need not lose any sleep. By the way, at a $50,000. premium level I'd like to be part of that ins. consortium.
Its probably fair to say some percentage of contestants have less than an equal chance of winning; but at this level, I'd say most of the remainder are pretty evenly matched. That's to say, for instance, I don't believe any of the former winners of this contest would have a 40-1 chance of winning this year. Assuming 396 entrants I would say any one of their chances are much better than 396-1, but not close to 40-1 (or 100-1). Assuming an 80% chance that Player X is entered into both contests would reduce odds to 50x and 125x, respectively. That's reasonable. The better scenario would be for one of GE's insurance subsidiaries to underwrite the risk at cost, then we're talking a range of just $8,000-$20,000. [/QUOTE]I doubt 50% would enter both contests. I'll be surprised if it's over $12K.

One could easily provide the data from four NFFC years and 6-7 WCOFF years and look at where the winners of one year have finished all the other years and would determine the odds are remote. [/QUOTE]From an underwriting/actuarial view this would be insignificant. Ignoring even the fact that the sample size isn't meaningful, it doesn't matter what someone did year to year as the payout would be for a single year's performance. I wouldn't be surprised if KJ was close with his inital analysis for the first year.

As to the number of common entrants, the fact that you COULD win a million if you entered both events would lead to more entering both events, which is the whole idea behind offering the prize, right?

[RiFF]

Originally posted by 4D:
[
As to the number of common entrants, the fact that you COULD win a million if you entered both events would lead to more entering both events, which is the whole idea behind offering the prize, right? Agreed, that's how I get to a 50% overlap, without the bonus incentive I do not believe it approaches 50%
And actually, I believe the objective of the incentive is marketing and name recognition. This should get a lot of press and make a much larger audience aware of NFFC/NBC than would otherwise be aware of these contests.