Diamond Prize Fight Score Threshold Mountain (Data Analysis)

[ghostR]

So in the past two weeks, I have finally gotten strong enough to get into the Diamond Prize Fight. Great timing I know.

Before we get into things let's just explain that the changes really started at the BIG BAND Prize Fight

Prize Fight		Score Difference	TOP 10%	Score Difference	Percentage Increase
	TOP 30%	TOP 30%	TOP 10%	TOP 10%
Filia (Start of Rotation	32,644,683		50,585,245
Parasoul	31,247,467	-1,397,216	50,413,595	-171,650	-0.34%
Beowulf	35,046,350	3,798,883	50,013,822	-399,773	-0.79%
Peacock	35,380,400	334,050	50,078,097	64,275	0.13%
Painwheel	31,619,158	-3,761,242	46,842,914	-3,235,183	-6.46%
Double	34,557,474	2,938,316	50,356,626	3,513,712	7.50%
Squigly	36,256,518	1,699,044	50,195,556	-161,070	-0.32%
Valentine	34,299,893	-1,956,625	46,319,774	-3,875,782	-7.72%
Fukua	37,978,729	3,678,836	49,566,396	3,246,622	7.01%
Eliza	39,427,794	1,449,065	51,819,728	2,253,332	4.55%
Annie	35,866,314	-3,561,480	48,328,695	-3,491,033	-6.74%
Big Band	41,776,049	5,909,735	54,095,843	5,767,148	11.93%
Cerebella	44,928,675	3,152,626	58,117,266	4,021,423	7.43%
Robo-Fortune	47,540,211	2,611,536	62,302,743	4,185,477	7.20%
Ms. Fortune	49,511,635	1,971,424	64,921,807	2,619,064	4.20%
Fillia (Not Official - End of Rotation)	55,000,000	5,488,365	70,000,000	6,078,193	9.36%
				TOTAL INCREASE	36.95%

Honestly wish this form took spreadsheet colours.

Rotation 1 vs Rotation 2
	TOP 30% THEN	TOP 30% NOW	DIFFERENCE	TOP 10% THEN	TOP 10% NOW	DIFFERENCE
Filia (Start of Rotation)	32,407,828	32,644,683	236,855	46,028,987	50,585,245	4,556,258
Parasoul	32,201,684	31,247,467	-954,217	43,934,593	50,413,595	6,479,002
Beowulf	34,727,927	35,046,350	318,423	46,419,068	50,013,822	3,594,754
Peacock	34,325,520	35,380,400	1,054,880	45,301,580	50,078,097	4,776,517
Painwheel	31,969,492	31,619,158	-350,334	45,187,830	46,842,914	1,655,084
Double	34,616,691	34,557,474	-59,217	44,998,567	50,356,626	5,358,059
Squigly	36,491,443	36,256,518	-234,925	47,294,305	50,195,556	2,901,251
Valentine	29,197,786	34,299,893	5,102,107	43,629,613	46,319,774	2,690,161
Fukua	39,073,098	37,978,729	-1,094,369	46,055,589	49,566,396	3,510,807
Eliza	40,430,910	39,427,794	-1,003,116	48,499,060	51,819,728	3,320,668
Annie	38,450,558	35,866,314	-2,584,244	49,155,781	48,328,695	-827,086
Big Band	40,707,260	41,776,049	1,068,789	50,651,291	54,095,843	3,444,552
Cerebella	32,603,607	44,928,675	12,325,068	51,601,082	58,117,266	6,516,184
Robo-Fortune	28,781,954	47,540,211	18,758,257	50,141,775	62,302,743	12,160,968
Ms. Fortune	30,184,357	49,511,635	19,327,278	49,089,653	64,921,807	15,832,154

So what does all this basic data mean?

The exponential growth of scores from the BIG BAND Prize Fight. Since the changes and at the start of said prize fight, there has been exponential growth, while we haven't come full rotation to know just where it will go, it seems to be getting rapidly higher for both TOP 30% and TOP 10%.

The Robo-Fortune is a great example of growth. Previously, you had to get around 30 million to secure a top 30% spot, which wasn't too hard. Now to secure TOP 30% you would have needed to reach 50 Million to be safe. That is a whole 20 Million difference! When aiming for TOP 10% you had to reach around 54 million to be safe, but now you have to reach over 65 million and counting.

Where will this take us? Only the next post from @Sairus to see what the past few prize fights and weekend brings us. Only once we have done a full rotation will we know exactly how big of a change Diamond Prizefight has become.

I wont get into all the maths involved, but the recent 5 prize fights have revealed an increase on average of 4,534,261 score. This means that top 10% for the next prize fight would be around 70 Million (example from the chat given that the scores aren't out) + the average score increase of 4,534,261 = 74,534,261. Somewhere around there. Whereas top 30% has seen an average recent increase of 3,826,737.

Estimated Top 10% is 74,534,261+
Estimated Top 30% is 53,826,737+

DISCLAIMER: These are just REALLY rough calculations and guesses given the current times, trends and changes. Since we don't have the most recent score results. Would be happy to actually build a whole data capture for all the prize fights, trends and such recorded if Cellsai is willing to throw data at me.

Happy Grinding!

 

[evilgordo]

Well, it sorta becomes a self-fulfilling prophecy; after one falls short of the current PF threshold, everyone goes 4-5M more to be 'safe'.
But I also think the lock-out contributes in some way as thresholds are determined by percentile of the qualifying players, especially in Diamond.

Before the lock out, players could and did play in all 3 tiers but if a player wasn't equipped to handle say Robo's Gold modifier or not have the time, they would bail, but if they scored enough, their participation would add to the percentage pool. But now the pool is split into 4 separate pools and I think the Dia PF is the shallowest 'pool' of players, but with the most concentration of high scorers. If more (all) players were allowed to try the Dia PF, as long as they scored more than 1M (3 milestone rewards), there would be more data points and could shift the percentile thresholds down* (and in this case, most of us are only grinding for the 30%, which no one cared about Pre-Dia PF, it was always 10%).

*the percentile door swings both ways; it also possible around the BigBand PF cited was where many frustrated Dia PF players that kept missing the 30% threshold, went back to Gold PF, which would also raise scores as there are less players in the data pool)

 

[BallotBoxer]

what a beautiful board of data

I was going to make a "The PF Rewards System Has Always Been Bad" thread, but I think this does a good job illustrating the point. Rewards dependent on other players' actions outside of your knowledge or control isn't healthy competition. It just leads to bitter disappointment.

The last 4 PFs were crazy. I expected the scores to dampen as everyone divided their attention between normal PFs and the special Halloween Prize Fight. But it looks it is like evilgordo describes, less participants means less low scores to sink the average. Instead, the supreme players with dozens of level 60 diamond encrusted collections are soaring around with their usual half-a-billion scores, raising the rates.

I don't have the time nor the strength to keep up, so I've gone into cruise control until Umbrella arrives. Get the 12m milestone and bounce. Pointless to try to guess where 30% will end up. Guaranteed it will always be one million higher than where you stop.

 

[Codlugarthia]

Great job with the tracking thank you

I think its time to look at Milestone rewards rather than percentiles if the cutoff scores keep rising. That said scoring ramps up really quickly in Dia PF so the other key factor is time taken. Getting 65M now in Dia PF still feels like it's quicker than grinding top 10% in the Old Gold. I sometimes hit 2-3M points from a single fight now when theres a mulitplier bounty on the opponents team. They can be VERY tasty indeed.

 

[ZeoW]

Seems like not many people are aware of this but here's an excel sheet that is regularly updated and contains ALL data of Prize Fight and Rift cutoffs, including the ones since years ago:

SGM Score Cutoffs

docs.google.com

 

[ghostR]

Seems like not many people are aware of this but here's an excel sheet that is regularly updated and contains ALL data of Prize Fight and Rift cutoffs, including the ones since years ago:

SGM Score Cutoffs

docs.google.com

Well... There goes the rest of my weekend thanks. I live in data so there is that, glad to see people have been keeping track.

I knew about it but didn't really do much digging as it was a quick overview of the recent prize fight trends.

 

[evilgordo]

Sooooo....

Ok, the first 2 PFs results, post-Costume Party, are in. And interestingly enough, both Diamond 30% thresholds are under 30M!
While the 10% scores remained relatively close to the previous 70M range. (It's obvious the only factor at 10% is the new rewards? are they really that much better?)

The 30% is interesting, many factors could have contributed to the decline: Obviously the big difference is no Holiday PF, players that don't have a large, strong roster tend to focus one PF with the 'better' rewards. So, post-holiday PF, what happened? Was there that many players that only focused on the Holiday PF suddenly returned to Dia PF? Did the discouraged players that missed the previous Dia cutoffs go to Gold PF? Was it the characters? I'm guessing Beo/Pea were not super desirable as fighters/fodder? Probably a mix of many factors.

I'm guessing the 30% Dia thresholds now is only a 'soft reset' and will eventually get back to the mid-high 40M's as word gets out that scores are achievable again. But Seasons Beatings is right around the corner... another opportunity to see if the pattern repeats and maybe reassess how one will approach Dia PFs during that timeframe...which in turn becomes a self-fulfilling prophecy again...