Do teams with lower state tax rates get significant discounts?

[Dekes For Days]

I’m talking about what you initially did, which was dismissing the results of correlation in favor of dichotomizing two continuous variables and then averaging the averages of them. From that point on, you firmly entrenched yourself in a glass house and you’ve been throwing stones out of it ever since.

So you're talking about something that intentionally ignores everything that has happened since...

The only one throwing stones in a glass house is you. You posted a so-called analysis, based entirely on a separate multi-thread argument you've been engaging in, using data that you even admit is clearly incredibly flawed. You pretend that you had no pre-conceived ideas about the results, despite that being a blatant lie that is easily seen in pages and pages of your post history. You came to conclusions based on this flawed data that even your data doesn't support, with no justification besides you wanting it to say that.

When I tried to show the difference between the high-tax and no-tax areas that have always been debated (that was the very cause for this thread), you took issue with my methods and my sample (despite you literally using the average of an average above, despite the data being naturally grouped, and despite you separating the 5 no-tax areas in your own evaluation). So I literally switched my methods and sample to exactly match yours and what you said you wanted, and when they still showed the significant difference between high-tax and low-tax areas (over 2 different years), you completely dismissed my methodology and results based on equating it to an extreme example that you know full well doesn't apply here. You talk about extreme negative correlation examples with cherry-picked cut-offs to dismiss everything you don't like, despite your data literally showing a positive correlation, despite my results matching pretty closely with that correlation (as you've admitted), despite me using two different cut offs including the one you specifically chose, and despite those two differently sized samples showing the exact same thing over multiple years (that as I showed, would have exposed the lack of correlation in your provided example).

Instead of answering to any of it, you accuse me and belittle me and the results. You talk about things from the beginning of the thread, even though I literally changed everything to the way you wanted to do it. You still complain and accuse me and personally attack me, because you wanted to craft your narrative in here without opposition. Even if you want to believe that it's not perfect, your methods are far from perfect as well, so you're in no position to be outright dismissing things and criticizing others in the way that you have.

The results I showed match perfectly with all of the data we have, and it provides valuable information, but you want to dismiss it outright based entirely on hypotheticals that you know don't apply.

 

[TomasHertlsRooster]

So you're talking about something that intentionally ignores everything that has happened since...

The only one throwing stones in a glass house is you. You posted a so-called analysis, based entirely on a separate multi-thread argument you've been engaging in, using data that you even admit is clearly incredibly flawed. You pretend that you had no pre-conceived ideas about the results, despite that being a blatant lie that is easily seen in pages and pages of your post history. You came to conclusions based on this flawed data that even your data doesn't support, with no justification besides you wanting it to say that.

When I tried to show the difference between the high-tax and no-tax areas that have always been debated (that was the very cause for this thread), you took issue with my methods and my sample (despite you literally using the average of an average above, despite the data being naturally grouped, and despite you separating the 5 no-tax areas in your own evaluation). So I literally switched my methods and sample to exactly match yours and what you said you wanted, and when they still showed the significant difference between high-tax and low-tax areas (over 2 different years), you completely dismissed my methodology and results based on equating it to an extreme example that you know full well doesn't apply here. You talk about extreme negative correlation examples with cherry-picked cut-offs to dismiss everything you don't like, despite your data literally showing a positive correlation, despite my results matching pretty closely with that correlation (as you've admitted), despite me using two different cut offs including the one you specifically chose, and despite those two differently sized samples showing the exact same thing over multiple years (that as I showed, would have exposed the lack of correlation in your provided example).

Instead of answering to any of it, you accuse me and belittle me and the results. You talk about things from the beginning of the thread, even though I literally changed everything to the way you wanted to do it. You still complain and accuse me and personally attack me, because you wanted to craft your narrative in here without opposition. Even if you want to believe that it's not perfect, your methods are far from perfect as well, so you're in no position to be outright dismissing things and criticizing others in the way that you have.

The results I showed match perfectly with all of the data we have, and it provides valuable information, but you want to dismiss it outright based entirely on hypotheticals that you know don't apply.

Okay. Keep on averaging averages and dichotomizing continuous variables and then acting like you’re the smartest guy in the room. I’m sure you’ll learn a lot that way.

 

[WesMcCauley]

All of the tax discussion needs to get turned down a good amount. Does it have some influence? Sure. Is it enough to make it close to as big of a deal it might seem to be with all the talks about it? No.

NYR have no problem attracting free agents and its expensive as hell to live there, but people wanna sign there because they get to live NY. There are pros and cons to all markets when it comes to taxes, signing bonuses, cost of living, endorsement deals, weather, chances of winning, how the franchise is run (hallo Melnyk) etc. There are so many factors in play that just using taxes is a little stupid... Specially when many players find ways to pay less taxes than they probably "should" by having a house in another city and getting paid most of their salary in signing bonuses etc.

And btw, the salary cap was mostly for the owners to have cost certainty, not league parity.

 

[Dekes For Days]

Okay. Keep on averaging averages and dichotomizing continuous variables and then acting like you’re the smartest guy in the room. I’m sure you’ll learn a lot that way.

Once again, you're ignoring that you did the exact same thing, you're ignoring the countless other flaws in your own evaluations and data (that would get you laughed out of a high school math class), and you're trying to hide everything that has happened in this thread since, because it fits your narrative. I literally changed around everything in my evaluation to match what you were doing, and what you said you wanted, and every time it doesn't show what you want, you change what you want, complain, talk about the past, personally attack me, and try to outright dismiss the results using non-applicable examples, while you simultaneously use your incredibly flawed data and methods to reach unsupported, pre-conceived conclusions.

The only one pretending that they are the smartest one in the room is you. The data, as flawed as it is, is clear. There is a significant difference in what low-tax and high-tax areas pay.

 

[TomasHertlsRooster]

Once again, you're ignoring that you did the exact same thing, you're ignoring the countless other flaws in your own evaluations and data (that would get you laughed out of a high school math class), and you're trying to hide everything that has happened since, because it fits your narrative. I literally changed around everything in my evaluation to match what you were doing, and what you said you wanted, and every time it doesn't show what you want, you complain, talk about the past, personally attack me, and try to outright dismiss the results using non-applicable examples, while you simultaneously use your incredibly flawed data and methods to reach unsupported, pre-conceived conclusions.

The only one pretending that they are the smartest one in the room is you. The data, as flawed as it is, is clear. There is a significant difference in what low-tax and high-tax areas pay.

This coming from the guy who posted this in response to a thread based on correlation:

For the record, even by your evaluation, the average of the 5 no-tax states is 98.38%. The average of the 5 locations with taxes over 50% is 113.16%.

EDIT: A bigger analysis with a bigger sample of teams using OP's own methods of cap hit/projected cap hit:

2019:

All teams between 40-45% taxes: 100.4%
All teams between 49-53% taxes: 105.2%

2018:

All teams between 40-45% taxes: 101.7%
All teams between 49-53% taxes: 109.3%

Still a significant difference.

Do you not know when to just stop? And give up? You have made a complete fool of yourself and I have no interest in discussing this with you.

 

[Dekes For Days]

This coming from the guy who posted this in response to a thread based on correlation:

Yep, thank you for posting it again so that everybody can see. I posted a quick 2-minute evaluation, using your provided data, of the no-tax and high-tax areas that are always discussed, to show the significant difference (contrary to your conclusion that appears to be based on nothing). You had an issue with my methods, so I improved on them and did it in the exact way that you did (as seen below), adding up all of the individual cap hits and projected cap hits, while using the sample that you wanted.

The averages of the no-tax/high-tax areas, the in depth look at that same sample in 2019 (that you did), the in-depth look at that same sample in 2018 (that you did), the in-depth look at the expanded sample (that you chose) in 2019, the in-depth look at the expanded sample (that you chose) in 2018, and your correlation and graphs (that you did) all show the same thing. There is a significant difference between what low-tax and high-tax areas pay.

For the record, even by your evaluation, the average of the 5 no-tax states is 98.38%. The average of the 5 locations with taxes over 50% is 113.16%.

Cap Hit/Projected in 2019:
DAL, FLA, NSH, TBL, VGK: 99.07% of projected cap hit.
ANA, LAK, MTL, OTT, SJS, TOR: 108.95% of projected cap hit.

Cap Hit/Projected in 2018:
DAL, FLA, NSH, TBL, VGK: 100.34% of projected cap hit.
ANA, LAK, MTL, OTT, SJS, TOR combined: 108.62% of projected cap hit.

A bigger analysis with a bigger sample of teams (chosen by OP) using OP's own methods of cap hit/projected cap hit:

2019:
All teams between 40-45% taxes: 100.4%
All teams between 49-53% taxes: 105.2%

2018:
All teams between 40-45% taxes: 101.7%
All teams between 49-53% taxes: 109.3%

View attachment 248967

View attachment 249215

 

[Strakanator]

Two of the highest state/local tax areas are NYC and California. Those areas still seem to attract a good amount of free agents.

The important thing to remember is that tax planning & strategy often can mitigate a good amount of a locations advantage over another.

Let’s say the NY plus NYC highest tax rate is 13%. Assuming no bonus, 50% of the players contract would be subject to the NY/NYC tax for home games, 6.5%. The effective tax rate would get cut down based on the brackets and tax planning, which could adjust the rate down to around 3-4%. It’s not that big of an advantage.

If a player gets a huge bonus, then the player is usually in a higher federal tax bracket that year and pays more federal tax as opposed to state tax. This could result in little or no net tax savings.

If a player wants the most money in their hands today, then the player will negotiate a huge bonus, pick a no income tax state, and pay the highest U.S. federal tax rate, 37% on all income over $500k. That is a killer percentage of tax being paid for anyone in the tax biz.

The lowest tax rates will be achieved by taking the income over years and using tax planning to reduce the effective rate every year.

I am not sure players or agents really understand it. They just want the most money today. That seems to be the trend the league is heading in.

A time-value of money analysis assumes the markets will trend upwards. At best, the markets will most likely produce a lower rate of gain than in the past.

In conclusion, there is more value in non-front loaded contracts than the players probably are aware.

 

[ThunderRoad]

The assumptions in the source data are not qualified from what I could tell. How was team tax rate determined - did it take into account all local taxes, sales tax, property tax (and housing insurance costs), etc? Personally when I moved back to Florida, any advantage of no state income tax was completely eliminated by the higher property tax/insurance cost factored into my mortgage (actually costs more to live in Florida than state moved from that had a state income tax). There are a lot of factors that go into true cost of living in an area (and median salary for that area) more so than just state income tax or lack thereof. As well as all the other factors others mention like team success, endorsement/cultural opportunities, weather, family considerations, etc. Contracts aren't negotiated simply based on team tax environment, so there will always be bias when comparing player salaries from different teams or even within a team.

(As an example that the source data consists of inaccuracies the state I moved from back to Florida has about a 4.5% higher tax rate associated with it, and I can guarantee that is not true considering all the other taxes/higher cost of goods/higher utility/service costs pay now living in FL than in that state - more expensive now where looking at those numbers in a vacuum would suggest I am saving).

Without knowing many of the assumptions used when manipulating the data, hard to know the true bias introduced. And the correlation/matrix plot show no discernable correlation at all (random scatter) with coefficient more or less zero. Props for undertaking the analysis though.

 

[TomasHertlsRooster]

The assumptions in the source data are not qualified from what I could tell. How was team tax rate determined - did it take into account all local taxes, sales tax, property tax (and housing insurance costs), etc? Personally when I moved back to Florida, any advantage of no state income tax was completely eliminated by the higher property tax/insurance cost factored into my mortgage (actually costs more to live in Florida than state moved from that had a state income tax). There are a lot of factors that go into true cost of living in an area (and median salary for that area) more so than just state income tax or lack thereof. As well as all the other factors others mention like team success, endorsement/cultural opportunities, weather, family considerations, etc. Contracts aren't negotiated simply based on team tax environment, so there will always be bias when comparing player salaries from different teams or even within a team.

Without knowing many of the assumptions used when manipulating the data, hard to know the true bias introduced. And the correlation/matrix plot show no discernable correlation at all (random scatter) with coefficient more or less zero. Props for undertaking the analysis though.

This is where I got the "estimated tax rate" information from.

Income Tax Calculator - CapFriendly - NHL Salary Caps

Also, can you elaborate on what you mean by the assumptions used when manipulating the data, and the true bias?

Yep, thank you for posting it again so that everybody can see. I posted a quick 2-minute evaluation, using your provided data, of the no-tax and high-tax areas that are always discussed, to show the significant difference (contrary to your conclusion that appears to be based on nothing). You had an issue with my methods, so I improved on them and did it in the exact way that you did (as seen below), adding up all of the individual cap hits and projected cap hits, while using the sample that you wanted.

The averages of the no-tax/high-tax areas, the in depth look at that same sample in 2019 (that you did), the in-depth look at that same sample in 2018 (that you did), the in-depth look at the expanded sample (that you chose) in 2019, the in-depth look at the expanded sample (that you chose) in 2018, and your correlation and graphs (that you did) all show the same thing. There is a significant difference between what low-tax and high-tax areas pay.

Yes, I looked at things the same way that you did, for the sake of humoring you, after you brought up the top-5/bottom-5 thing; I just showed you the better way to do things, instead of averaging averages. I did not initially bring up the top-5/bottom-5 thing, but merely added a quick note about the bottom-5 teams in the OP. However, I also made it very clear that testing for correlation is much better than dichotomizing continuous variables that do not contain the entire sample, even if you are not testing averages between them and I explained to you why. Your "bigger analysis with bigger sample" quite literally excluded teams from the sample in the OP.

Again, I don't think you understand what a significant correlation is. The r^2 is less than 0.1 in both years In both years, the p-value is considerably greater than 0.1, which means that the correlation here is literally not significantly significant, even at the 90% confidence level. If a confidence level is not stated, it is generally assumed that we are talking about a confidence level of 95%. The fact that you still can't seem to grasp this, yet tell me that I would get laughed out of a high school math class, and that I have an agenda, really bugs me. You are the one with an agenda here. You are saying that data is significant when it is quite literally not even statistically significant at the 90% confidence level.

 

[Dekes For Days]

after you brought up the top-5/bottom-5 thing

You brought up the bottom 5 thing in your opening post, and the top-5/bottom-5 thing is literally why this entire thread was created.

I just showed you the better way to do things

Showed me the "better way to do things", and then when it didn't show you what you wanted, suddenly it's 100% completely worthless and worthy of ridicule.

merely added a quick note about the bottom-5 teams in the OP.

A "quick note" that presented the information in an incredibly manipulative way, which you then drew conclusions from.

However, I also made it very clear that testing for correlation is much better than dichotomizing continuous variables that do not contain the entire sample, even if you are not testing averages between them and I explained to you why.

No you didn't. You posted an extreme outlier example that doesn't even apply, and used it to dismiss any piece of information that wasn't yours (even some that were yours). This despite literally every piece of information we have (based on your data) saying the exact same thing, that there is a significant difference between what low-tax and high-tax areas pay.

Your "bigger analysis with bigger sample" quite literally excluded teams from the sample in the OP.

Once again, it was your sample that you wanted, and it only excluded the teams in the center of the tax percentage range. After recalculating, I actually think I made a mistake and put the 2019 40-45% percentage higher than it should have been, helping your case. But if you want to look at the entire sample, fine:

All contracts between 40-41% taxes over 2 years: 99.7% of projected cap hit
All contracts between 41-45% taxes over 2 years: 100.9% of projected cap hit
All contracts between 45-49% taxes over 2 years: 106.4% of projected cap hit
All contracts between 49-53% taxes over 2 years: 107.5% of projected cap hit

Still a significant difference between low-tax areas and high-tax areas, and we see a consistent trend.

Again, I don't think you understand what a significant correlation is.

You have an incredibly weak and incomplete data set with a ton of issues, so you aren't going to see perfect correlation, and you know that. The fact that we see the correlation we do, and these clear trends in the data, speaks volumes.

 

[ziggyjoe212]

Most players favor quality of life more than taxes. For example, NYC will always be a major UFA player, whereas Columbus tends to struggle to sign UFA's.
The tax difference is not an issue for most players.

 

[txpd]

And btw, the salary cap was mostly for the owners to have cost certainty, not league parity.

That's not really true at all. There were a number of teams like the Capitals that felt like they had no realistic chance to compete in an open salary league. So, the cap was also about parity.

 

[thestonedkoala]

Just less than 700.
Around the Don Mills area

Fair enough, but it seems that the average income in Toronto is much higher than in Austin.

 

[Krewe]

Because the wealthiest fan bases have to pay money for teams like Florida and Arizona to have teams?

If you had to pay extra for your neighbour to drive a Lexus at half price while you drove a Honda. That would probably bug you as well

Yes you personally pay for those places to have teams

Of course they do Stamkos at 8.5 mil, Kucherov at 9.5 mill, Karlsson at 5.9 and BARKOV at less than 6 million is you need, not to mention your boys Meier and Labanc.

It doesn't happen in 100% of cases but it happens more often than not

If you are going to try and take a shot at other fans, at least don't be totally wrong when you do. They are literally playing in the highest tax area in the nation.

Also Stamkos was MAYBE worth 9 when he signed, and Kucherov was probably worth 10-10.5. Considering they resigned with their original team, I would say those are reasonable hometown discounts

 

[TomasHertlsRooster]

You brought up the bottom 5 thing in your opening post, and the top-5/bottom-5 thing is literally why this entire thread was created.


Showed me the "better way to do things", and then when it didn't show you what you wanted, suddenly it's 100% completely worthless and worthy of ridicule.


A "quick note" that presented the information in an incredibly manipulative way, which you then drew conclusions from.


No you didn't. You posted an extreme outlier example that doesn't even apply, and used it to dismiss any piece of information that wasn't yours (even some that were yours). This despite literally every piece of information we have (based on your data) saying the exact same thing, that there is a significant difference between what low-tax and high-tax areas pay.


Once again, it was your sample that you wanted, and it only excluded the teams in the center of the tax percentage range. After recalculating, I actually think I made a mistake and put the 2019 40-45% percentage higher than it should have been, helping your case. But if you want to look at the entire sample, fine:

All contracts between 40-41% taxes over 2 years: 99.7% of projected cap hit
All contracts between 41-45% taxes over 2 years: 100.9% of projected cap hit
All contracts between 45-49% taxes over 2 years: 106.4% of projected cap hit
All contracts between 49-53% taxes over 2 years: 107.5% of projected cap hit

Still a significant difference between low-tax areas and high-tax areas, and we see a consistent trend.


You have an incredibly weak and incomplete data set with a ton of issues, so you aren't going to see perfect correlation, and you know that. The fact that we see the correlation we do, and these clear trends in the data, speaks volumes.

So you think we should draw a conclusion based on data with a p-value greater than 0.1 and an R^2 less than 0.1? Okay.

 

[Legion34]

Okay. Keep on averaging averages and dichotomizing continuous variables and then acting like you’re the smartest guy in the room. I’m sure you’ll learn a lot that way.

• Goes on for pages in unrelated thread denying low tax market have advantages
• Thread has multiple quotes and projections from agents, managers tax specialists proving they DO sign players for less
• Makes a thread ignoring the actual evidence from pros claiming that this is fan opinion and Confirmation bias and IGNORES actual evidence from pros. To try to be the smartest in the room
• Pretend has no agenda and no idea how will go.
• Ignores actual contracts and actual comparisons from High and low tax markets in favour of estimated contracts. That NEVER happened.
• Even in this method purposely ignores multiple variables and designs a “study” that supresses effects.
• STILL finds the relationship that was told existed the entire time.
• Just decided this doesn’t count and makes conclusions that don’t follow from ACTUAL evidence or the HYPOTHETICAL evidence from his own data.

Then you Complain when other people use your own data to prove the point that actual people who sign Actual contracts have said all along

 

[Dekes For Days]

So you think we should draw a conclusion based on data with a p-value greater than 0.1 and an R^2 less than 0.1? Okay.

You are literally the one that came here with this data, posted it, and started drawing conclusions. I am merely saying that based on your data, your conclusions are incorrect.

You're taking maybe 5% of the data (if you're lucky), which includes contracts for players of differing ages, positions, term, abilities, contract statuses, signing conditions, etc. over a limited time period, comparing it to an untested and admittedly flawed projection model, and then plotting the differences. You created a data set filled with giant swings and inconsistencies from individual contracts, that was designed to have weaker than average correlation, and then have used that weaker correlation to claim that the data says something it does not, while ignoring all context. You know that the positive correlation we do have over multiple years, even under these conditions, speaks volumes.

All of the data we have clearly shows a significant difference between high-tax and low-tax areas, with a clear, consistent trend throughout the sample. I even showed you this, including the whole sample as you requested, and you have nothing to say except to misrepresent my position and the situation.

 

[TomasHertlsRooster]

You are literally the one that came here with this data, posted it, and started drawing conclusions. I am merely saying that based on your data, your conclusions are incorrect.

You're taking maybe 5% of the data (if you're lucky), which includes contracts for players of differing ages, positions, term, abilities, contract statuses, signing conditions, etc. over a limited time period, comparing it to an untested and admittedly flawed projection model, and then plotting the differences. You created a data set filled with giant swings and inconsistencies from individual contracts, that was designed to have weaker than average correlation, and then have used that weaker correlation to claim that the data says something it does not. You know that the positive correlation we do have over multiple years, even under these conditions, speaks volumes.

All of the data we have clearly shows a significant difference between high-tax and low-tax areas, with a clear, consistent trend throughout the sample. I even showed you this, including the whole sample as you requested, and you have nothing to say except to misrepresent my position and the situation.

There is literally not a statistically significant correlation between the two variables. That is stats 101.

Please, ditch the attitude and take a stats class. I’ve literally never seen anybody so adamant, arrogant, and smug about something they are factually incorrect about.

 

[Dekes For Days]

There is literally not a statistically significant correlation between the two variables. That is stats 101.

You would be told in Stats 101 that having such an incomplete data set that doesn't even cover all of the teams, let alone contracts, would be wrong.
You would be told in Stats 101 that basing your entire data set on an untested projection model would be wrong.
You would be told in Stats 101 that drawing the conclusions that you did based entirely on "weak" correlation that was designed to be weak would be wrong, especially when the correlation is positive and consistent across multiple years.
You would be told in Stats 101 that there are valuable methods of evaluation other than using (heavily manipulated) correlation coefficients in isolation.

You shouldn't be making any conclusions based on this data. But if that's what we're doing, then the data points to a very clear result; that there is a significant difference in what low-tax and high-tax areas pay. Which matches what pretty much everybody in the industry says.

 

[TomasHertlsRooster]

You would be told in Stats 101 that having such an incomplete data set that doesn't even cover all of the teams, let alone contracts, would be wrong.
You would be told in Stats 101 that basing your entire data set on an untested projection model would be wrong.
You would be told in Stats 101 that drawing the conclusions that you did based entirely on "weak" correlation that was designed to be weak would be wrong, especially when the correlation is positive and consistent across multiple years.
You would be told in Stats 101 that there are valuable methods of evaluation other than using (heavily manipulated) correlation coefficients in isolation.

You shouldn't be making any conclusions based on this data. But if that's what we're doing, then the data points to a very clear result; that there is a significant difference in what low-tax and high-tax areas pay. Which matches what pretty much everybody in the industry says.

Is that what you learned in your stats class, where they also taught you to dichotomize continuous variables and compare the averages of averages?

 

[Dekes For Days]

Is that what you learned in your stats class, where they also taught you to dichotomize continuous variables and compare the averages of averages?

Ah, I see you've moved back to the "I have no answer so I'll just personally attack you and deflect and misrepresent everything" portion of this discussion.

 

[TomP24684]

How many cups can we all agree that our maple leafs would have if the system wasn’t completely rigged against us?

 

[TomasHertlsRooster]

Ah, I see you've moved back to the "I have no answer so I'll just personally attack you and deflect and misrepresent everything" portion of this discussion.

There is literally nothing to discuss here.

You are dismissing the results of a regression which shows a slight correlation that is present twice, but is not statistically significant, in favor of results that you obtained from dichotomizing two continuous variables. I am not sure what else there is to discuss?

 

[Legion34]

Ah, I see you've moved back to the "I have no answer so I'll just personally attack you and deflect and misrepresent everything" portion of this discussion.

I think what was learned was

• Ignore actual first hand accounts of negotiations and actual tax specialists
• Ignore ACTUAL signed contracts in the league in high and low tax markets. That would be easy to do. He did it for pages. In the last thread.
• Instead compare signed contracts and hypothetical contracts that don’t even exist
• Find a correlation despite doing everything you can not to
• Make conclusions about it .....
• but when it turns out the initial prediction is wrong and others prove EXACTLY what you were told all along despite efforts to suppress effects
• Resort to insults and name calling

That sounds about right

 

[Dekes For Days]

There is literally nothing to discuss here.

You're right. There isn't. You have attempted to manipulate the information and misrepresent the results, and you have failed. You have attempted to hide the countless flaws and problems in the methods you use, while ignoring, attacking, misrepresenting, and/or outright dismissing everybody else's methods (even when they construct it exactly as you want), and you have failed. You have attempted to manufacture unsupported conclusions and push a narrative, and you have failed.

The data, while incredibly flawed, is extremely clear and consistent across multiple years and methods (even more than would have been expected given the problematic data set), and matches what the industry says. There is a significant difference between what low-tax markets and high-tax markets pay.