merchandising

[CardinalsFan]

tysok,

thats why everything is "editable"!

i agree with your sentiments though. i interpret merchandising as also money from ballpark vendors, parking, etc in addition to uniform sales, etc. i gave every team in my league a baseline of $20million and then bumped it up for some successful teams. worked like a charm as now the difference is 1.7 to 1 versus the original 50 to 1. the only bad thing is i will need to revisit this every 5 years or so, but i am happy so far in the yearly adjustments.

[scefalu]

Without question, merchandising as it stands is broken.

Tysok: I agree with you that market size should be a function of population, not initial payroll. However, that's a different issue.

CardinalsFan: I think your system could serve as the basis for a reworked merchandising scheme. Teams with high fan loyalty (e.g., New York Yankees and Chicago Cubs) will always have high sales in branded gear. On the other hand, the only time you'll ever see Brewers jerseys flying off the shelves (barring earthquakes) is when they are in the playoffs.

If I had my way, merchandising revenue would hinge on three factors:

1. Market Size - sets the baseline for the merchandising revenue
2. Fan Interest - adds to the baseline because everybody loves a winner and wants to wear their hats
3. Fan Loyalty - high fan loyalty should make declines in the merchandising revenue slower as the fans are more willing to tolerate failure (see Cubs)

Now, there can be bumps for big free agent signings and blockbuster trades becuase everyone needs to get new jerseys with Boof Bonser's name and number (Love that name!). But I think the 3 factors listed above are the significant ones.

Suggestions? Improvements? Thoughts?

[tysok]

Originally Posted by scefalu

Tysok: I agree with you that market size should be a function of population, not initial payroll. However, that's a different issue.

It's a slightly different issue. It's not a completely different issue since merchandising doesn't change very rapidly, and is driven by market size... which is the start of the whole problem... but anyhow...

Let's take a look at some arbitrary numbers. I'll use the last year I've played in my main league as the basis for numbers from the game.
Attendance in the game is fairly equivalent to reality. 75 million fans went to see games this year, in my league 80 million did. The top team in reality, the Yankees, drew 4.2 million fans. in my game the top team drew 3.8 million. The worst, the Marlins drew 1.1 million, my worst was 1 million.
So gate Revenue works out fairly well.

Media revenue, which is also based on market size I have a slight issue with. A lot would have to go into this to make it work out better, it can't just be market size. The team in my game that only managed to draw 1 million fans has a media contract of 25 million, while the team that drew 3 million has 13 million. Since fan interest should be a volatile number, fan loyalty should have a base in that... The 2nd worst team in attendance has the 4th highest media contract, they have a market size of 8 but a loyalty of 4. The team came in last in attendance has a market size of 4 and loyalty of 6... they received the 24th largest media contract. Both are perennial losers, 3 or few playoff appearances in 20+ years. But still, that isn't the biggest killer... it's a minor annoyance that could be a result of the end of a contract and their next one will be much bigger, or smaller... harder to know than merchandising.

Payrolls, which is the result of all the revenue that comes in. There's way too much money floating around the league in the game. In reality there's only 5 teams that broke that 100 million mark. In the game I have 9 teams that have. In reality only 2 teams are over 104 million, in the game there are 8. 3 teams are over 128 million in the game, to 1 in reality. That's not as bad as it gets... there are 18 teams in the game that have a budget of over 100 million... and 10 of those are over 150 million. With all that money out there it creates an inflation type effect as the teams spend their money poorly.

Revenue sharing, which should level the playing field a little more, doesn't work right. It's a government welfare program. Teams don't get money from it, they get debt forgiveness from it. Those that make over the cash limit lose all their money, those that are negative in cash get enough money to make it to 0. In my league 33 million dollars disappeared, it was taken but not distributed. It should be taken (probably from the teams it took it from) and distributed on a scale to every other team, which means the worst team may go over the cash max. The problem is these teams are so pathetic they're treading water, they constantly lose money and are constantly brought up to 0, but never given enough to jump in to the competition and raise their revenues so they become self sufficient. It needs to become a revenue sharing system intended to help energize the teams, not a government welfare program intended for debt forgiveness.

Revenue Sharing is a problem, but merchandising revenue is the biggest killer, and merchandising is what we're talking about here.
I agree that merchandising revenue isn't just the sale of jerseys, it's not just how much is made by selling a shirt with Boof Bonser's name. It's vendors and parking and anything else the team does to make money. The problem is the game doesn't differentiate. It gives the Twins all the money from selling Bonser's jersey, and all the money for it's vendor sales etc.
Let's take a look at attendance in the game, and some arbitrary numbers to come up with merchandising revenue.
Say everytime a fan came to the game they drove one extra person with them. So (Attendance / 2) * 7 dollars for parking.
Say every single fan spent 5 dollars on trinket merchandise to vendors. So Attendance * 5.
Say every single fan spent 8 dollars on food. So Attendance * 8.
The highest attendance team in my league would then receive 69 million dollars for merchandising. The worst team, drew only 1 million fans, would make 19 million.
Now we should take something into account for the price of tickets... if the best team was charging 18 dollars a ticket, while the worst was chargin 12, the worst teams fans have more money to spend on the merchandise. I wouldn't weight this to strongly since it wouldn't be a direct correlation... But it should have a little something built into that.
Let's ignore that though and look at how my estimates compare with the numbers from the game.
The best team in attendance in my game made 27 million in merchandise. The worst made 17 million. Those aren't the best or worst in the league. The best merchandise revenue in the game is 74 million dollars... they were 12th in attendance. The worst merchandise revenue in the game was 3 million, they were 18th in attendance.
In one instance every single fan that went to a game spent 51 dollars on merchandise, that was the 29th team in attendance. The team with 3 million in merchandise revenue averaged every single fan spending 1 dollar on merchandise. That's just not right.
A range of 69 to 19 is a whole lot better and more realistic than a range of 74 to 3. A team that has won enough to get their fan interest high enough to draw more fans to the park should be reaping more money from those fans coming to the park... with exceptions of how high they drive their ticket prices up.

Going by speculation, the game is saying that (in my game) MLB sold 90 million in merchandise, split it out and gave 3 million to each team, and that lousy team sold absolutely nothing, and gave free parking, to the 1 million fans that came to their park.
Since merchandising revenue should consist of only what you sold at your park (plus some amount that the league sold and split among all teams) the merchandising revenue number HAS to be relative to attendance.
Market size should only come into play in how many fans you may get to attend. Fan loyalty and interest comes into play with the media contracts and attendance.
That's how it should be... right now it's all mixed up together in some undiscernable way.

[scefalu]

Tysok: I agree that merchandising should be related to attendance. You say attendance, I say fan interest. Since fan interest directly influences the attendance, I think we're going along the same path on this one.

It seems that the initial draft sets the payroll, which determines the market size. With that comes a set media contract. Now, some semi-random attendance figure is assigned based on the (payroll-based) market size. Finally, the game assigns a merchandising value so that every team has profit. The magnitude of the profit seems to vary along with (you guessed it) the payroll, with the largest market teams ($110+M payroll) making a $60M profit from the previous year. The only way to do this is assign $90+M in merchandising revenues. At the same time, the smallest market team (which drew 2.26M fans?) gets a measly $1.25M in merchandising.

Putting it into per-fan spending ... the large market fans spent $36.50 each in addition to their tickets while the small market fans spent about $0.55 each.

HOORAY FOR CHEAP BEER!

My point ... that's too wide a range. (I think that goes along with CardinalsFan and Tysok.)

Okay, fine so merchandising isn't really merchandising, it is the rug that OOTP sweeps its financial dirt under. But the merchandising revenue never changes. If it changed at some point I could say, "that's OOTP for you!"

I know that Markus and Co. don't want the front office to be the focus of the game, but this is too big of a distraction. Simply leveling the merchandising curve by an order of magnitude, or so, would help a lot. That way, even if the values never change, there isn't a built-in $89M advantage for some teams.

(Okay, I'm done ranting.)

[battists]

Originally Posted by scefalu

I know that Markus and Co. don't want the front office to be the focus of the game, but this is too big of a distraction. Simply leveling the merchandising curve by an order of magnitude, or so, would help a lot.

Well, like I said Shawn, come up with a model that is more graceful, but without requiring a degree in economics, and I'll pitch it for you. We know now that it ain't going to change in 2006, so let's come up with a good idea for how it could work in 2007.

[CardinalsFan]

Originally Posted by battists

Well, like I said Shawn, come up with a model that is more graceful, but without requiring a degree in economics, and I'll pitch it for you. We know now that it ain't going to change in 2006, so let's come up with a good idea for how it could work in 2007.

ahh a challenge. i have a weeks vacation next week to come up with something.

[fhomess]

Tysok, I'm not sure where you get the idea that your numbers are realistic or not, which is not to say that they aren't, but we should look at real data instead of guessing, and you didn't quote a source for any of your numbers. Anyway, you prompted me to do some research since I think you're thinking along the right lines...

In 2001, the Giants pulled in just over $61 million in local revenue with 3.3 million fans, which does not include game receipts, broadcasting, or post-season income. The Expos pulled in just under $3 million with about 650k fans (the Marlins were around $4 million with about 1.2 million).
You can find the revenue numbers (courtesy of the late Doug Pappas) for 2001 here: http://roadsidephotos.sabr.org/baseball/mlbsez.htm
I pulled attendance from Baseball-Reference.com.

So Giants fans spent about $18.48 in local revenue per attendee, while Marlins fans spent $3.33 per attendee (Expos were at $4.30-ish). I would be surprised if the gap hasn't actually increased since 2001.


It seems there's two sources of merchandising revenue:
1. Items purchased at the stadium, not including tickets. We'll call this local revenue.
2. Items purchased away from the stadium. We'll call this national revenue.

National revenue is shared equally among all teams. When MLB reports national revenue, it includes both broadcasting and merchandising revenue. In 2001, all MLB teams got $24.4 million in national revenue, with Arizona and Tampa Bay being exceptions who got slightly less.

As for local revenue, you can get data about ticket prices and Fan Cost Index (FCI - the average cost for a family of 4 to attend a ball game), at the same website I quoted earlier in this post.

It seems to me that Fan Cost Index - 4*Average Ticket price is a reflection of the average amount a fan would spend at a game. Now, not all fans will purchase everything in the FCI, but some would purchase more. I suppose it might depend on fan interest or fan loyalty. The ratio of FCI to average ticket price from 1992 to 2002 ranged roughly between 7 and 13, with most teams being between 8 and 10.5.

Anyway, back to the 2001 numbers for Montreal, Florida, and San Francisco. I'll throw in the Yankees, too:

Team           Attendance    Local Revenue    Rev/Att   Avg Tix       FCI    FCI-4*Tix
Florida         1,200,000      $ 4,037,000     $ 3.36    $14.37   $117.60       $60.12
Montreal          650,000      $ 2,829,000     $ 4.35    $ 9.70   $ 80.08       $41.28
New York (A)    3,260,000      $47,057,000     $14.43    $28.90   $192.60       $77.00
San Francisco   3,312,000      $61,524,000     $18.58    $23.38   $181.02       $87.50

If I take the Rev/Attendee as a percentage of FCI-4*Tix, then that gives me an idea of fan interest. So, in the example of Florida, they got $3.36/$60.12 = 5.6%. Montreal got 10.5%, New York got 18.7%, and San Fran got 21.2%. So it seems that a fan interest weighted measure of how much of the FCI came back might be a good approach, with teams realizing between 5% and 25% of their FCI as per fan merchandising revenue.

FCI = Avg Ticket Price * (Fan Interest weighted number between 7 and 13)
Merchandising Revenue Per Fan (MRPF) = (FCI - 4*Avg Ticket Price) *(Fan interest weighted number between .05 and .25)
Total Merch Revenue (TMR) = Attendance * MRPF

So in Tysok's example, we had the worst team charging $12 and the best team charging $18 per ticket. If we say the best team had a fan interest of 100, and the worst team had a fan interest of 25, and we're using a linear scale between 7 and 13 (I'd probably use a quadratic, but that's me). Let's also assume that the best team drew 4 million fans, and the worst team drew 1 million.

Team18 FCI = $18*(1.00*(13-7)+7) = $234
Team18 MRPF = [$234 - 4*($18)] * [1.00*(.25-.05) + .05] = $40.50
Team18 TMR = $40.50 * 4,000,000 = $162,000,000

Team12 FCI = $12*(0.25*(13-7)+7) = $102
Team12 MRPF = [$102 - 4*($12)] * [0.25*(.25-.05) + .05] = $5.40
Team12 TMR = $5.40 * 1,000,000 = $5,400,000

What if we narrow the range for FCI factor from 7-13 to 8-10.5:

Team18 FCI = $18*(1.00*(10.5-8)+8) = $189
Team18 MRPF = [$189 - 4*($18)] * [1.00*(.25-.05) + .05] = $29.25
Team18 TMR = $29.25 * 4,000,000 = $117,000,000

Team12 FCI = $12*(0.25*(10.5-8)+8) = $103.50
Team12 MRPF = [$103.50 - 4*($12)] * [0.25*(.25-.05) + .05] = $5.55
Team12 TMR = $5.55 * 1,000,000 = $5,550,000

So we've now instead of a difference of $162 million to $5.4 million, we've got $117 million to $5.55 million, which is clearly still too much, but you can kind of get the idea from here. Let's reduce the high end by taking our 25% factor and lowering that to 15%.

Team18 FCI = $18*(1.00*(10.5-8)+8) = $189
Team18 MRPF = [$189 - 4*($18)] * [1.00*(.15-.05) + .05] = $17.55
Team18 TMR = $17.55 * 4,000,000 = $70,200,000

Team12 FCI = $12*(0.25*(10.5-8)+8) = $103.50
Team12 MRPF = [$103.50 - 4*($12)] * [0.25*(.15-.05) + .05] = $5.55
Team12 TMR = $5.55 * 1,000,000 = $5,550,000

That won't affect the low end team, but the high end team is now down to $70 million. Actually, if you take that $17.55 and multiply it by the 3.3 million fans San Fran actually had, you're now at $58 million, which is less than what they actually made. If you wanted to get more complicated, you could make this non-linear, but this is simple and takes only three lines to code. The only variables between teams are fan interest as a percentage and ticket price. The FCI rate range (e.g.: 7-13, 8-10.5) coud be adjusted easily with a variable in the code, as could the MRPF percentage range (e.g.: .05-.25, .05-.15).

[scefalu]

Originally Posted by battists

Well, like I said Shawn, come up with a model that is more graceful, but without requiring a degree in economics, and I'll pitch it for you. We know now that it ain't going to change in 2006, so let's come up with a good idea for how it could work in 2007.

As you wish ... I have sent you a function to pitch.

Edit: It looks like this (see attached figure).

A - Base value for the function
M - Market size (1-9)
L - fan loyalty (1-10)
I - fan interest (1-100)
n - Exponent on market size (I recommend 0.5 or less)
C1 - Sets the lower bound for merchandise income
C2 - Weights the fan loyalty
C3 - Weights the fan interest

This function can be tuned to give the curve shape you want and provide a changing value of the merchandise revenue from year to year.

By setting A=100,000,000; n=0.001; C1=0.6; C2=C3=0.2; I am able to get a function that tracks pretty closely with the attendance (fan interest), but with enough variation form the loyalty and market size.

[scefalu]

fhomess: Nice information and breakdown. Thanks for the hard work. I like the fact that the spread between teams can be on the order of $70M. It makes me feel a little better about the spread I'm seeing in some fictional leagues. It addresses all my concerns. I wonder how hard it would be to implement.

[battists]

Originally Posted by scefalu

As you wish ... I have sent you a function to pitch.

Care to share it with the peeps before I pitch it?

[scefalu]

Originally Posted by battists

Care to share it with the peeps before I pitch it?

Fixed in the edit above.

[scefalu]

And here's the function results ...

The data is the old and new (proposed) merchandising revenue for a 16 team league (2 sub-leagues of 2 divisions, 4 teams per division) over 11 simulated seasons.

The bottom feeders in the attendance pool are getting a LOT of merchandising revenue. (see fhomess's post above) I justify this as a replacement for revenue sharing. The function can be tuned, so the spread between the max and min can be made to suit your tastes and the exponent can be used to control the curvature.

[battists]

Scefalu,

Let's assume that in 2007, Markus is able to fix revenue sharing so that it more closely resembles the "real world" version.

If that happens, is your function still valid? It sounded almost like you had "corrected for" revenue sharing in there. Ideally, whatever I give to Markus should be "the way it SHOULD work," and not "the way it COULD work if revenue sharing is broken".

Let me know!

Steve

[scefalu]

Okay, change C2 and C3 to 0.5 ... so that teams with really low loyalty and interest have nobody spending extra money at their parks, and you get results that are similar to what fhomess got with his calculations.

This is how it should work if it is in a revenue-only mode.

[tysok]

Fhomess: I was looking for that type of information for a while and couldn't find it. Most of what I said was examples, except the attendance figures.

I really like that formula too. Everything checks out and it puts up good numbers with real game data. The final representation I think is the better one to go with, taking the high end from 25% to 15%. It keeps things in line a little bit better.
Here's the data I get from this equation (at 25%) with the league I was using as my base:

Tickets	Interest	 Att. 	Rev by Formula	 Rev by Game 	Difference
18	100	 3,835,008.00 	 112,173,984.00 	 27,601,236.00 	84,572,748.00
18	97	 3,725,485.00 	 105,127,971.02 	 15,201,756.00 	89,926,215.02
18	100	 3,140,259.00 	 91,852,575.75 	 30,499,740.00 	61,352,835.75
17	87	 3,724,119.00 	 87,570,423.81 	 18,329,814.00 	69,240,609.81
17	85	 3,715,016.00 	 85,101,729.02 	 13,956,624.00 	71,145,105.02
17	92	 3,420,461.00 	 85,721,541.31 	 60,898,230.00 	24,823,311.31
17	89	 3,140,799.00 	 75,781,512.35 	 31,384,584.00 	44,396,928.35
17	81	 3,391,310.00 	 73,639,244.47 	 27,694,548.00 	45,944,696.47
17	82	 3,333,892.00 	 73,378,629.53 	 5,620,124.00 	67,758,505.53
17	81	 3,328,380.00 	 72,272,776.16 	 65,289,564.00 	6,983,212.16
17	84	 3,202,620.00 	 72,400,349.29 	 6,538,314.00 	65,862,035.29
17	81	 3,111,874.00 	 67,571,543.22 	 74,192,274.00 	-6,620,730.78
17	92	 2,736,426.00 	 68,578,666.56 	 3,103,272.00 	65,475,394.56
17	89	 2,797,615.00 	 67,501,134.48 	 23,174,262.00 	44,326,872.48
17	87	 2,818,498.00 	 66,275,289.37 	 17,315,370.00 	48,959,919.37
15	80	 3,077,068.00 	 58,156,585.20 	 26,227,314.00 	31,929,271.20
15	76	 2,958,699.00 	 52,892,662.02 	 26,013,960.00 	26,878,702.02
15	75	 2,984,099.00 	 52,594,744.88 	 27,702,972.00 	24,891,772.88
15	65	 2,641,259.00 	 40,114,121.06 	 17,608,104.00 	22,506,017.06
15	64	 2,240,314.00 	 33,497,174.93 	 22,999,950.00 	10,497,224.93
15	63	 1,998,605.00 	 29,415,468.39 	 20,620,332.00 	8,795,136.39
15	53	 2,231,126.00 	 27,800,945.52 	 17,500,050.00 	10,300,895.52
15	64	 1,804,505.00 	 26,980,958.76 	 53,867,106.00 	-26,886,147.24
15	59	 1,911,293.00 	 26,370,109.52 	 19,649,466.00 	6,720,643.52
15	52	 1,885,773.00 	 23,087,518.84 	 25,814,376.00 	-2,726,857.16
15	50	 1,832,446.00 	 21,645,768.38 	 27,536,598.00 	-5,890,829.63
15	47	 1,460,448.00 	 16,324,887.74 	 4,586,901.00 	11,737,986.74
15	49	 1,294,617.00 	 15,016,909.89 	 66,189,150.00 	-51,172,240.11
14	30	 1,472,122.00 	 10,768,572.43 	 3,355,397.00 	7,413,175.43
15	40	 1,010,669.00 	 9,854,022.75 	 17,500,050.00 	-7,646,027.25

It's taking merchandise revenue away from those teams that can't draw fans, as it should be, and putting more in to those that actually play in front of people. It takes the range from 74 down to 3 and makes it 112 to 10 though... bigger range and not warranted quite that much I don't think. Plus, as you can see, it messes up my columns... which is not to be tolerated.

Here's the data if the formula goes to 15%:

Tickets	Interest	 Att. 	Rev by Formula	 Rev by Game 	Difference
18	100	 3,835,008.00 	 67,304,390.40 	 27,601,236.00 	39,703,154.40
18	97	 3,725,485.00 	 63,335,294.02 	 15,201,756.00 	48,133,538.02
18	100	 3,140,259.00 	 55,111,545.45 	 30,499,740.00 	24,611,805.45
17	87	 3,724,119.00 	 53,558,696.71 	 18,329,814.00 	35,228,882.71
17	85	 3,715,016.00 	 52,221,515.54 	 13,956,624.00 	38,264,891.54
17	92	 3,420,461.00 	 52,019,054.98 	 60,898,230.00 	-8,879,175.02
17	89	 3,140,799.00 	 46,200,132.53 	 31,384,584.00 	14,815,548.53
17	81	 3,391,310.00 	 45,503,495.40 	 27,694,548.00 	17,808,947.40
17	82	 3,333,892.00 	 45,261,584.57 	 5,620,124.00 	39,641,460.57
17	81	 3,328,380.00 	 44,659,121.12 	 65,289,564.00 	-20,630,442.88
17	84	 3,202,620.00 	 44,502,967.00 	 6,538,314.00 	37,964,653.00
17	81	 3,111,874.00 	 41,754,113.97 	 74,192,274.00 	-32,438,160.03
17	92	 2,736,426.00 	 41,616,113.89 	 3,103,272.00 	38,512,841.89
17	89	 2,797,615.00 	 41,152,007.43 	 23,174,262.00 	17,977,745.43
17	87	 2,818,498.00 	 40,534,440.37 	 17,315,370.00 	23,219,070.37
15	80	 3,077,068.00 	 36,001,695.60 	 26,227,314.00 	9,774,381.60
15	76	 2,958,699.00 	 32,992,452.55 	 26,013,960.00 	6,978,492.55
15	75	 2,984,099.00 	 32,871,715.55 	 27,702,972.00 	5,168,743.55
15	65	 2,641,259.00 	 25,628,466.23 	 17,608,104.00 	8,020,362.23
15	64	 2,240,314.00 	 21,453,246.86 	 22,999,950.00 	-1,546,703.14
15	63	 1,998,605.00 	 18,886,067.77 	 20,620,332.00 	-1,734,264.23
15	53	 2,231,126.00 	 18,355,752.49 	 17,500,050.00 	855,702.49
15	64	 1,804,505.00 	 17,279,939.88 	 53,867,106.00 	-36,587,166.12
15	59	 1,911,293.00 	 17,109,178.20 	 19,649,466.00 	-2,540,287.80
15	52	 1,885,773.00 	 15,291,733.26 	 25,814,376.00 	-10,522,642.74
15	50	 1,832,446.00 	 14,430,512.25 	 27,536,598.00 	-13,106,085.75
15	47	 1,460,448.00 	 10,996,625.77 	 4,586,901.00 	6,409,724.77
15	49	 1,294,617.00 	 10,045,095.13 	 66,189,150.00 	-56,144,054.87
14	30	 1,472,122.00 	 7,831,689.04 	 3,355,397.00 	4,476,292.04
15	40	 1,010,669.00 	 6,822,015.75 	 17,500,050.00 	-10,678,034.25

The same type of result. It's taking the money away from those that it should, and giving more to those who earned it. It's taking the range from 74 down to 3 and making it 67 to 7... a much better range in my opinion. And it doesn't mess up my columns... a huge bonus!

You see that merchandising revenue of 66 million, the 2nd worst in attendance, and the 2nd highest revenue in the majors... that's ridiculous. And although you can't see it, I'll tell you they earned their low attendance.

This is only one step in the process... but it's a big one and a good one.
This formula takes into account the ticket price. That's good, it works with the fact that people have to spend money to get there and may not spend more.
It takes into account fan interest. The more interest there is the more you may sell.
It takes fan loyalty into account in a round about way, if you have a high fan loyalty more people will come to the game even when interest wanes, allowing ticket prices to stay higher maybe... or at least upping the attendance in general. But doesn't give any weight to loyalty in particular, which loyalty is useless to a team if rear ends aren't in the seats.
And finally it takes attendance as a whole into account. It's a beautiful piece of work that deals straight out of reality, which makes it even better.

If this were implemented, revenue sharing worked correctly, and the media contracts were tweaked, I think the whole financial system would have a much better outlook.

Fhomess, would you have time to look at media revenue (if you can find some data on it) and see how it works out? Seems like it's off in the game, related more to market size than to fan interest than it maybe should... but I don't know. It seems like it so I keep saying it. The Yanks probably have the highest real media revenue, but I'm fairly sure that if nobody cared about the Yankees they wouldn't have the highest revenue just because they're in the biggest market.

I'd love to see this formula be used for merchandising... it's simple, it's realistic, and most importantly it works right.

[fhomess]

Scefalu's equation takes into account market size and fan loyalty, which mine does not. It dawned on me as I was thinking about this last night that for my Fan Cost Index to be based on Fan Interest doesn't really make much sense. In real life, the FCI doesn't change during the regular season. It's a measure of how much certain items (4 tickets, parking, a program, food, and a souvenir or two) would cost a fan at that ballpark. Teams don't change these prices during the year. For realism's sake, the FCI should be calculated using fan loyalty instead, as that doesn't really change during a season either, and is rather steady over the course of many seasons.

Here's how mine would look, then:

TIX   - Average Ticket Price
FL    - Fan Loyalty
FCI   - Fan Cost Index
FCIh  - High End FCI Factor
FCIl  - Low End FCI Factor
FI    - Fan Interest
MRPF  - Merchandising Revenue Per Fan
MRPFh - High End MRPF Percentage
MRPFl - Low End MRPF Percentage
ATT   - Attendance

FCI = TIX * [(FL/10) * (FCIh-FCIl) + FCIl]
MRPF = (FCI - 4*TIX) * [(FI/100) * (MRPFh-MRPFl) + MRPFl]

Total Merchandising Revenue = MRPF * ATT

Recommended Values:
FCIh  = 10.50
FCIl  =  8.00
MRPFh =  0.15
MRPFl =  0.05

As for how market size affects this, I think scefalu's equation puts too much emphasis on it. In the 2001 data set I took my previous numbers from, market size seems to be pretty random in terms of how it affects the local revenue:

Team               Local Revenue*
San Francisco        $61,524,000
Seattle              $56,211,000
New York (AL)        $47,057,000
Cleveland            $45,295,000
Los Angeles          $41,100,000
New York (NL)        $38,162,000
Atlanta              $37,692,000
Milwaukee            $37,010,000
Houston              $36,826,000
Colorado             $35,197,000
Texas                $34,561,000
Arizona              $32,970,000
Chicago (NL)         $30,642,000
Baltimore            $29,691,000
Boston               $29,485,000
Tampa Bay            $28,633,000
St. Louis            $27,581,000
Pittsburgh           $26,598,000
Chicago (AL)         $26,291,000
Anaheim              $26,195,000
Detroit              $21,018,000
Toronto              $14,255,000
Oakland              $13,932,000
Kansas City          $13,270,000
San Diego             $8,504,122
Philadelphia          $7,739,000
Minnesota             $6,987,000
Cincinnati            $6,523,000
Florida               $4,037,000
Montreal              $2,829,000

*Does not include gate receipts, local TV, radio, or cable revenue.

On the other hand, local broadcasting revenue does seem to be tied somewhat to market size. Here's a table of those teams' local broadcasting revenue, as well as where they ranked in the previous table of merchandising revenue:

Team             Local TV, Radio, Cable    Rank of Local Revenue
New York (AL)         $56,750,000                   3
New York (NL)         $46,251,000                   6
Seattle               $37,860,000                   2
Boston                $33,353,000                  15
Chicago (AL)          $30,092,000                  19
Los Angeles           $27,342,000                   5
Texas                 $25,284,000                  11
Chicago (NL)          $23,559,000                  13
Cleveland             $21,076,000                   4
Baltimore             $20,994,000                  14
Atlanta               $19,988,000                   7
Detroit               $19,073,000                  21
Philadelphia          $18,940,000                  26
Colorado              $18,200,000                  10
San Francisco         $17,197,000                   1
Tampa Bay             $15,511,000                  16
Florida               $15,353,000                  29
Toronto               $14,460,000                  22
Arizona               $14,174,000                  12
Houston               $13,722,000                   9
San Diego             $12,435,878                  25
St. Louis             $11,905,000                  17
Anaheim               $10,927,000                  20
Oakland                $9,458,000                  23
Pittsburgh             $9,097,000                  18
Cincinnati             $7,861,000                  28
Minnesota              $7,273,000                  27
Kansas City            $6,505,000                  24
Milwaukee              $5,918,000                   8
Montreal                 $536,000                  30

It's definitely weigthed to the top, but a closer look at that chart would cause you to realize it's not as simple as market size. Anaheim's rather low while LA is high up. No wonder they keep changing their name to get the folks from LA to feel like it's their team. I think the other effect here is team success. I've seen a few studies that show that team attendance is tied directly to team success. In Veeck As In Wreck, Bill Veeck, who was an owner of several major and minor league franchises including the Brewers, Indians, Browns, and White Sox mentions this repeatedly.

It seems logical to reason that the number of fans interested in watching your team at home is related to the number of fans that would take the time to actually come out to the ballpark. I haven't actually seen any studies of this, though. Anyway, the conclusion from that would be that team success would help you get a bigger contract out of your market than you otherwise might.

But the fact that the Mets and Yankees are so far beyond everyone else suggests to me that it is very largely related to market size. I don't know for sure, but I think Seattle gets high up on the list because of the connection with Japan (through Ichiro) and the fact that they've got a rather large regional market.

Wikipedia has a rather detailed article on this history of broadcasting contracts here: http://en.wikipedia.org/wiki/Major_L...sion_contracts, but this focuses more on national than local broadcasting.

One thing I will say for all of this, is that it'd be nice to find some more data, as basing everything on 2001 is a rather limited scope.

[scefalu]

I like the work fhomess has done. So, I wanted to see how his formula would work for the simulations I have been using for this discussion. I plotted the two merchandising revenue functions side-by-side and you can see how they compare in the figure below.

My settings are as posted previously. For fhomess's function, I used his recommended settings from the previous post.

First off, I agree that my function might place too much emphasis on market size. My rationale for this is that, in big markets, people who don't go to the games will still buy the home team's merchandise. As a result, my function appears as too generous to the big market bottom feeders. fhomess's function has a great track with the attendance, which we have all been looking for, and it rewards teams for higher attendance.

Looking at all this data so far. I would have to say that both functions are heavily dependent on picking the right fitting parameters. But I think the shape of fhomess's function is better.

[tysok]

Fhomess:

Looking at that media information I'd have to say it has a lot more to do with market size than I thought it would. St. Louis being so low is just odd, they're definately not small market, not big market, huge fan loyalty, and one of the lower media revenues.... Guess I'll stop mentioning that then.

About the merchandising revision... I can see the point. It actually works out in an understandable manner as well that I also like. I reran both sets of numbers (I didn't save the stuff I did last night, and I did another dump that replaced information earlier today) and this is what I get:

TIX	Loyal	Int	Att	 Loyalty Base 	 Interest Base 	Difference
18	10	100	3850217	 67,571,308.35 	 67,571,308.35 	0.00
17	8	89	3755539	 53,246,031.94 	 55,242,758.14 	-1,996,726.20
17	5	87	3715243	 45,427,204.97 	 53,431,045.85 	-8,003,840.88
17	7	87	3673709	 49,197,392.50 	 52,833,721.51 	-3,636,329.01
17	10	84	3572551	 52,898,762.66 	 49,643,454.19 	3,255,308.47
15	8	80	3153320	 36,893,844.00 	 36,893,844.00 	0.00
18	10	100	3140198	 55,110,474.90 	 55,110,474.90 	0.00
17	8	87	3124634	 43,663,635.52 	 44,937,158.22 	-1,273,522.70
17	9	83	3119919	 44,088,355.37 	 42,853,881.42 	1,234,473.95
17	10	86	3065102	 46,062,352.86 	 43,582,072.32 	2,480,280.54
17	9	82	3041103	 42,651,469.58 	 41,286,622.55 	1,364,847.03
17	9	84	3033100	 43,183,761.25 	 42,147,350.98 	1,036,410.27
15	7	75	2982651	 32,156,706.09 	 32,855,764.92 	-699,058.83
15	10	78	2961825	 36,963,576.00 	 33,835,888.80 	3,127,687.20
15	8	70	2851809	 30,799,537.20 	 29,516,223.15 	1,283,314.05
17	10	88	2819534	 42,995,073.97 	 41,010,685.94 	1,984,388.03
17	7	84	2811748	 36,829,681.18 	 39,071,487.86 	-2,241,806.68
15	10	69	2788007	 32,347,851.22 	 28,490,992.03 	3,856,859.18
17	8	93	2738061	 39,937,357.75 	 42,100,631.29 	-2,163,273.54
15	7	68	2692391	 27,401,809.40 	 27,163,532.80 	238,276.60
15	10	63	2576414	 28,385,641.25 	 24,346,146.14 	4,039,495.10
15	9	54	2161143	 21,071,144.25 	 18,036,899.48 	3,034,244.77
15	10	49	2003830	 19,341,969.08 	 15,547,967.45 	3,794,001.63
15	7	42	1747730	 13,868,237.55 	 12,179,930.37 	1,688,307.18
14	1	30	1690443	 8,046,508.68 	 8,993,156.76 	-946,648.08
15	9	51	1681843	 15,924,950.91 	 13,440,658.56 	2,484,292.34
15	5	45	1540864	 11,527,588.80 	 11,253,122.40 	274,466.40
15	4	48	1356962	 9,973,670.70 	 10,372,617.53 	-398,946.83
15	4	42	1346866	 9,293,375.40 	 9,386,309.15 	-92,933.75
14	4	27	1198779	 6,461,418.81 	 6,041,426.59 	419,992.22

It makes sense. The most drastic change is it takes 8 million away from the team that has big interest right now but only marginal loyalty.

Either way, this is definately the way it needs to be done. Very nice work.

Edit: Also, if you look at both sets of numbers (the ones here and the ones I did last night) you'll see how the merchandising might change. For instance, the lowest revnue team last night was 6.8 mil, today it would be 6.4 ... their interest obviously dropped. The first dump (the information I was using last night) was 10/30/2021 in my game. The stuff from today is 10/30/2022 in my game.

[1998 Yankees]

Wow, a few days off from the boards and look at all this! Nice work, fhomess, scefalu, and tysok (alphabetical order), and thanks to battists for functioning as highly-involved intermediary. You know, it's fascinating to see how the design of a major aspect of the game could be influenced by the hard work of a few dedicated fans and the willingness of game designers to listen to them. As you can tell, I have nothing to add to this discussion beyond cheerleading, but know that I will be watching for the results and eating scefalu's popcorn in the meantime!

[battists]

Originally Posted by 1998 Yankees

Wow, a few days off from the boards and look at all this! Nice work, fhomess, scefalu, and tysok (alphabetical order), and thanks to battists for functioning as highly-involved intermediary. You know, it's fascinating to see how the design of a major aspect of the game could be influenced by the hard work of a few dedicated fans and the willingness of game designers to listen to them. As you can tell, I have nothing to add to this discussion beyond cheerleading, but know that I will be watching for the results and eating scefalu's popcorn in the meantime!

It doesn't actually turn into "design of a major aspect of the game" until I can convince Markus to do it.