Coming up with an objective way to rank chase days/seasons

[samuel stone]

Personally i love this, i've always tried to do similar things myself with much less success. Being primarily a Kansas chaser, I also love seeing so many amazing Kansas days which I've chased in the index, all the way from Harper and Mulvane 2004 to Chapman 2016!!

 

[Dan Robinson]

I ran June 17, 2010 and June 4, 2015 through the calculator, though since I did not chase either of those, I had to estimate many of the metrics from chase reports I saw online. If anyone is more familiar with those days and could enter the data to the best of your memory here, that would be helpful:

http://stormhighway.com/qi.php

The June 17, 2010 QI-S came out to 49. Simla's came out to 78, handily beating out Dodge City for the #1 spot. I would really like to get some confirmation on that day's metrics though instead of using estimates. I used the following specs for the Simla day:

6 photogenic tornadoes
15 total visible tornadoes
1 anticyclonic
90 minutes of photogenic tornadoes
120 minutes of good structure

Also, if any chase vets know of days prior to 2000 that have a shot at making the top 10, please let me know!

 

[JeremyS]

This weekend I'll make a web page with a javascript calculator that everyone can play with - both by manipulating the different multipliers and factors and entering data for chase events.



I didn't chase this event, but it should definitely rank high based on what I know about it. To rank the event and each storm, I would need:

Total daytime photogenic tornadoes.
Total tornadoes that were visible to chasers.
Total time that photogenic tornadoes were visible to chasers.
Was the lightning good (frequent/close/photogenic)?
Was structure good? How long did it last?
Storm speed in knots.
Terrain (good, mixed or poor)
Road network quality
How evident was the forecast?
Did the tornadoes cause significant damage, injuries or deaths?
Were there any rare features (two or more tornadoes at once, horizontal vortices, sunlit tornadoes, audible roars)?

My personal experience and somewhat reduced memory from nearly 8 years ago.... I've also attached a pdf from the Minneapolis WFO of the event. I am focused on the southern Minnesota tornadoes.
http://www.crh.noaa.gov/Image/mpx/StormReports/17June2010.pdf

This file lists 25 tornadoes in southern Minnesota all during the late afternoon and evening.
I see in the PDF, that 24 of the 25 tornadoes were videographed or photographed.
Shows tornadoes from 345-845pm.
I do remember a fair amount of lightning and a lot of those were the positive CG strikes.
Great structure. No issues with rain wrapped tornadoes, etc.
I remember storm speeds of 20-35 mph.
Terrain was also fairly good. Not a ton of trees, some hilly farmland, but other than that no real issues.
Road networks was fine.
Day 1 had a fairly large moderate risk with a 10% hatched tornado risk. Day 2 outlook for this day was a slight risk with a large 30% outlook.
Tornadoes did cause some significant damage to farmsteads. There was 1 fatality and a couple of injuries.
Tornadoes included 1 EF-4 and 3 EF-3's and 3 EF-2's. A couple of instances of 2 tornadoes on the ground at once. I witnessed horizontal vortices on the EF-4 tornado. The EF-4 was on the ground for 45 minutes, a mile wide at times, and threw a car 3200 feet!

 

[JeremyS]

I ran June 17, 2010 and June 4, 2015 through the calculator, though since I did not chase either of those, I had to estimate many of the metrics from chase reports I saw online. If anyone is more familiar with those days and could enter the data to the best of your memory here, that would be helpful:

http://stormhighway.com/qi.php

The June 17, 2010 QI-S came out to 49. Simla's came out to 78, handily beating out Dodge City for the #1 spot. I would really like to get some confirmation on that day's metrics though instead of using estimates. I used the following specs for the Simla day:

6 photogenic tornadoes
15 total visible tornadoes
1 anticyclonic
90 minutes of photogenic tornadoes
120 minutes of good structure

Also, if any chase vets know of days prior to 2000 that have a shot at making the top 10, please let me know!

I just made a post about the Minnesota tornadoes, not seeing you had made this one a few minutes earlier as I was typing up my reply!

 

[Dan Robinson]

Thanks for the info! The highest tornado count I saw from a chaser was 13 from Roger Hill, so I used that as the max for that southern target. I guessed 5 photogenic tornadoes on that storm based on the photos I saw from Roger's, Skip Talbot's and Jeff Duda's chase logs. That southern target had a base score of 56 when not counting for the fact that the F4 caused a death and injuries.

 

[Dean Baron]

Version 4 run:

View attachment 16704

v4 is now live on the calculator:

http://stormhighway.com/qi.php

Dan,

Just a small detail. Your date for the Dupree storm is wrong. It was on 6/16/10, the day before the big Minnesota outbreak on 6/17/10.

 

[Dan Robinson]

Version 5 updated:

QI-S = ((DPT*350)+(FL*50)+(H*400)+(TF*50)+(RF*50)+(R*50)+(TRF*-75)+(D*-75)+(I*-150)+(FT*-500)+(MD*-1500)+(L*50)+(T*50)+(HT*5)+(DPTmins*1.1)*(60-speed)+(Smins/30)*(60-speed))/100

Here are the rankings with the new tweaks and corrections along with a few more events added:



http://stormhighway.com/qi.php

 

[Chad M]

Where would 5/9/16 rank on the list ?

Definitely one of the days I regret missing, especially being somewhat close to home.

 

[Dan Robinson]

I added a few more events (including Eads and Katie) to the spreadsheet as well as updating the formula to "version 6" which increases the weight of all of the various factors to spread out the results between roughly 0 and 100:



In this new version, a score of 10 is "good", 20 is "excellent", 30 and up is "exceptional".

The calculator has also been updated:
http://stormhighway.com/qi.php

 

[JamesCaruso]

.... just three days come to mind that accounted for the majority of the tornadoes or otherwise noteworthy chase days: 5/1 (Culver), 5/27 (Wyoming) and 5/28 (Colorado landspouts).

Many saw tornadoes, but aside from 5/28, those events were very localized and fairly short-lived..

Quincy, thanks for the objective analysis. Nice to see quantified what we all anecdotally know to be true.

Question though, why do you consider the 5/28 Colorado landspouts event to NOT be a “localized and fairly short-lived event”? Maybe I’m just biased because I missed it, and admittedly I’m not sure how long it lasted (but doubt it was much longer than the WY 5/27 tornado), but wouldn’t you agree 5/27 it was most certainly a fairly localized mesoscale accident?

Also I think 5/29 in KS/OK is worthy of being included in the list of noteworthy chase days - not great, but at least as good as 5/27and 5/28.

 

[Dan Robinson]

Using the still-experimental Quality Index, the May 28 storm-level score is 57, which is higher than any ranked-to-this-point storm in 2017.

http://stormhighway.com/qi.php

May 28 won't score that high on the event-level, since only one target out of two/three performed tornado-wise, but it is still enough to at least get 2018 out of contention for worst season ever, at least by this metric. In 2006, there were no good events available during peak season (and I understand 1988-89 were similarly destitute).

 

[Ethan Schisler]

Unless I’m reading that chart wrong....I don’t see how May 28th can rank above Bowdle, Pilger, Cherokee OK, Carpenter, plus many other much more quality chase days. Maybe it’s just my opinion but all those aforementioned days were much greater of quality and intensity than May 28th both on a meteorological and chase scale. No disrespect I just was truly curious as to how that was calculated. Mods if this is off topic feel free to remove.

 

[Dan Robinson]

That's mainly because the scale takes points off for human impacts (deaths, injuries, damage etc) - the second column from the last is the value before the negative points are applied.

 

[Mikayla Norris]

I'm curious how huge outbreaks such as April 27th 2011 would stack up using this. Sure the shear number of tornadoes, the rare magnitude of the event, etc. may lend itself to a higher ranking, however the devastating cost in life and property would knock it down quite a bit as well. Come to think of it, I'm not even sure how feasible it would be to attempt to rank such huge, widespread events.

 

[Michael Gavan]

I've had a project going on Discord that roughly parallels this concept but its more of a 'here's all the data we have' and trying to subject-ify it as much as possible. Since my thread title would be too similar I think i'll just piggy back on to this one.

The questions we are trying to answer is A) What seasons are over/underrated. B) How Chase-able was X season. C) How do bad years compare to good by the numbers. D) What would storm chasing have been like in years like 1965, 1991, etc for those of us who weren't around, alive, or old enough to experience them.

Reduced to ONLY M-A-M-J.

One data set includes all plains states. MT WY ND SD CO NE MN IA IL IN KS OK NM TX

Another data set includes just KS N-TX and OK. I called this one the "Norman Index" The most typical chase territory of say, an OU student or someone who lives in the OKC metro specifically because of their hobby or job related to chasing.

From there, I wanted to compare quality, quantity, by days and by totals. I'll post some graphs. Attempt to upload the data. and then give a few parting thoughts.