Cactus Game Design Message Boards

People sometimes talk about trying to determine the best overall players in Redemption. Some say RNRS is a great system, some say that it favors those who can attend more tournaments. Some say it is the best option we have. However, I firmly believe that National tournaments are the strongest indicator of the skill level of players.

As a result, I did a lot of number crunching (by creating some macros that saved me hours of work) and came up with the top 50 players in T1 2p over the past 10 years (including active and retired players) by analyzing their finishes at the national tournament only.

As far as my spreadsheet - yes I will share it but I am still going to do some further work with the spreadsheet so I don't plan to make it available for everyone quite yet (I will eventually email it to all interested).

I filtered out all of the players who only played T1 2p at one national tournament over the past 10 years as I believe consistency is a necessary requirement to show your true skill level. Most of the top average finishers were those that only played in one national tournament.

The macro I created in Excel was based off the following formula:
For every year a player played at nationals, divide player's finish by the total number of players in that given year. The player's finish is thereby represented in terms of percentile. Add all of the percentiles together from each of the years that a player participated in the main event at nationals and divide by the number of years the player has played at nationals. Then multiply by the average number of players at national tournaments over the past 10 years (82.8 players).

As a result - the "average finish" is represented as the place the player would have obtained each year if there were 82.8 players at nationals each year.

Top Average T1 2p Finishes at Nationals Past 10 Years ('03-'12), min 2 Nationals - Active and Retired Players
1   Gabe Isbell   3.951150754
2   John Nesfeder   4.131757093
3   Daniel Huisinga   4.737346556
4   Matt Stupienski   5.419636364
5   Chad Soderstrom   5.92843956
6   Burton Reed   6.163252033
7   Tim Mierzejewski   6.727672328
8   Jonathan Greeson   7.06957043
9   Jeremiah Bryant   7.348780488
10   Matt Townsend   8.259255031
11   Joel Hey   9.054146341
12   Jonathan Alstad   11.34420906
13   Jordan Alstad   11.43043902
14   Roy Cannaday   11.61593496
15   Justin Sangillo   11.7974644
16   Jesse Epperson   12.39725275
17   Martin Miller   12.77632767
18   Nic Marshall   13.35608392
19   Ben Arp   13.59351577
20   Ken Shartle   13.60926829
21   Christian Fong   13.87637562
22   Daniel Whitten   14.48241758
23   Nick Archick   15.43090909
24   Kirk Dennison   16.2097689
25   Sara Harris   16.27128455
26   James Bryant   17.35658537
27   Nathan Voigt   17.41884749
28   Connor Magras   17.75585557
29   Josh Knitt   17.88662766
30   James Roepke   17.96954189
31   Luke McCrae   18.01105691
32   Mark Underwood   18.59912873
33   Josh Pearson   18.73091396
34   Drew Pegram   18.89365854
35   Timothy Reed   20.58780488
36   Christian Rohrer   20.75032753
37   Sam Nurge   21.1979095
38   Chris McCravy   21.61455052
39   Christian Kamke   21.83736264
40   Josh Ernst   21.8641785
41   Tim Maly   21.87938444
42   Jonathan Steckmann   22.10243902
43   Brian Jones   22.44828885
44   David Purvis   23.65409756
45   Alex Olijar   23.67014129
46   Josh Brinkman   24.01243328
47   Michael Wolfe   24.49120879
48   Jordan Drott   24.58809524
49   Jonathan Pequignot   25.16355875
50   Britta Alstad   25.22755401

This list shrinks considerably once we look at active players. I have identified only 28 of the 50 above as being "active" - and even some of those only play once every year or two (I am looking at you Justin Sangillo.)

Top Average T1 2p Finishes at Nationals Past 10 Years ('03-'12), min 2 Nationals - Active Players
1   Gabe Isbell   3.951150754
2   John Nesfeder   4.131757093
3   Daniel Huisinga   4.737346556
4   Matt Stupienski   5.419636364
5   Tim Mierzejewski   6.727672328
6   Jonathan Greeson   7.06957043
7   Matt Townsend   8.259255031
8   Jordan Alstad   11.43043902
9   Justin Sangillo   11.7974644
10   Martin Miller   12.77632767
11   Nic Marshall   13.35608392
12   Christian Fong   13.87637562
13   Nick Archick   15.43090909
14   Kirk Dennison   16.2097689
15   Nathan Voigt   17.41884749
16   Connor Magras   17.75585557
17   Josh Knitt   17.88662766
18   James Roepke   17.96954189
19   Mark Underwood   18.59912873
20   Sam Nurge   21.1979095
21   Tim Maly   21.87938444
22   Jonathan Steckmann   22.10243902
23   Brian Jones   22.44828885
24   Alex Olijar   23.67014129
25   Josh Brinkman   24.01243328
26   Michael Wolfe   24.49120879
27   Jonathan Pequignot   25.16355875
28   Britta Alstad   25.22755401

I hope y'all enjoy perusing this as much as I do. Now that I have put in 4+ hours of work into getting this far it will be easier going forward to put together new data of interest.

God bless,
Kirk

Interesting.

I knew I wouldn't be near the top as I've got a few bad years in there, including 2003 where I finished at the bottom in my first ever national tournament. But I would have thought my back to back 2nd's would have at least gotten me on the list.

It'd probably be helpful to include the number of played Nats in T12p within that timeframe next to the names.

In regards to it being weighted...quality over quantity.

Jonathan,
I am glad to help. 

John,
I will later play with some numbers and share pieces of trivia and play what if games such as removing your poor finish and Tim's poor finish.

I will add the number of tournaments played for each player and repost later.

To clarify, Daniel Huisinga has played in three national tournaments. He has placed in the top 10 in each of them including third place at Minnesota in 2005.

Jayden,
I will create a similar ranking list for T2 2p soon.

Kirk

Jonathan,
I am glad to help. 

John,
I will later play with some numbers and share pieces of trivia and play what if games such as removing your poor finish and Tim's poor finish.

Well, if you are removing poor finishes...erase my Ohio finish.      

This is why any ranking system will inherently fail, and why I would oppose using a ranking system to determine tournament seeds as Prof Underwood is proposing.

I agree that ranking systems are so hard to work - so many elements that cause the standings to change based on various criteria. RNRS ranking benefits those who can attend more tournaments and those who have weaker competition. 10 year nationals' rankings benefit those who consistently do well every time at nationals over those who did really well one year and poorly another year.

Realize that this was not the intent of Kirk's list. He was just number-crunching for the fun of it (call it curiosity). 

YMT,
Usually when it comes to numbers I have all sorts of ideas that pop in my head as to how I can use or think about the data. I like numbers and analysis.

My initial thought was actually to show a way you could determine the top 20 active players or so and them see how many of them were truly in the top 8 this past year at the point a top cut could have been used. I was curious how many of them would have missed the top cut and thus been able to provide a high level of competition for players such as those in your group who want to play against the best. Only 11 of the top 50 players on the active list even played T1 2p at nationals this year though. Interesting - I could make this more convoluted by weighting each national tournament based on this initial list - the more top players who attended the tournament, the more higher places matter at those tournaments.... (This is a difficult idea to work out and I don't know if I will get to that).

Another idea I had with this list - which will also take a lot of time - is to see what a top cut would have done to shake up the top 20 players at every nationals and then redo the calculations to show how many players have benefited from Swiss over the years and how many have been hurt.

John,

You are actually 56th with an average finish of 27.52547567. So close. 

But if it makes you feel any better - you are directly in front of Andrew Wester (27.67674005) and six spots in front of Chris Ericson   (29.58232625). 

Kirk

How do you take into account the fact that tournament size varies greatly?  After all a first at 2005 should (naively) count more than a first at 2012, given that 2005 had twice the number of players.

It would probably be better doing a rating system similar to chess (ELO) or X-box Live (modified GLICKO).  This would automatically include strength of schedule and tournament size and ... into the rankings. I can send anyone the code to calculate GLICKO, which I used previously for ROOT. In order to sue this code, you need all of the individual game results. Unfortunately, it takes a bit of time to strip these out of the spreadsheet, format them correctly, and make sure the names match from year to year.

39   Christian Kamke   21.83736264
40   Josh Ernst   21.8641785
41   Tim Maly   21.87938444

Hey Kirk,
Thanks for spending time with these rankings. In your 1st post on the 1st list, I copied the above^.
I know Christian since he's my son (I better know him haha) and I love him but I know he's not better than Maly overall. I know he beat him that game at Nats in Ohio, but overall accomplishments included, he's no Tim Maly. 

Perhaps not, but this system is slightly in favor of younger people. People like Malay, who were good back in the day, lose ranking in this system by continuing to play after they've gone to seed. The reason Gabe's at the top is that he's still at the top of his game, and the likely reason that Christian is above Malay is that he's yet to peak but still does well.

How do you get off the top of your game then?

How do you take into account the fact that tournament size varies greatly?  After all a first at 2005 should (naively) count more than a first at 2012, given that 2005 had twice the number of players.

Kirk already accounted for tournament sizes in his original calculation (see above).  However another thing that I'd like to see thrown into the formula is a factor to compensate for success in the current meta.  So basically a top finish 3 years ago would not be worth as much as a top finish this past year.  This would give a more up-to-date view of who the current top players are.

Clearly I'm not that good of a player I'm not on any of these lists - This takes so much pressure off

Here are the Top 25 Highest Value GLICKO Ratings from Nationals Only (2002-2012)

1. Jonathan Alstad           2067    2005
2. Keith Bartram               2031    2002
3. Christian Fong              2029    2008
4. Chad Soderstrom         2019    2007
5. Jeff Lau                        2018    2009
6. Tim Mierzejewski          2017    2008
7. Matt Stupienski            2002    2010
8. Allen Clark                   1988     2005
9. Tyler Stevens               1988    2010
10. Gabe Isbell                1986    2010
11. John Nesfeder           1981    2005
12. Ben Arp                     1962    2007
13. Josh Hey                   1960    2003
14. Nic Marshall               1954   2010
15. Daniel Huisinga         1953   2005
16. Brian Garlow             1945   2002
17. Justin Sangillo           1942   2005
18. Jonathan Greeson     1936  2007
19. Martin Miller               1935  2012
20. Chris McCravy            1934  2005
21. Tim Maly                     1929  2004
22. Jonathan Brodhacker 1912  2002
23. Josh Stanley               1889  2007
24. Kyle Hostetler            1879   2003
25. Justin Cannaday        1877   2003

Highest defiend to be the highest value at the end of a Nationals.  The third number is the year the ranking was achieved.
Starting values 1500 +- 350
All rankings carried over from year to year.

Where do I rank on these lists, EJB?