2018 Stock Market Indicator.

Metrics, Metrics, Metrics. In 2012 I produced my first Stock Market Indicator article and it has delivered solid results each year.
In 2012 Ohio St had a Stock Market Indicator (SMI) of +5.5, which would be a Bull Market. The +5.5 was calculated by taking their average win total from 2009-2010, which was 11.5 (11-2 in ’09, 12-1 in ’10) and then subtracting their wins from ‘11 (just 6) which produced the +5.5. The Buckeyes clearly underperformed in ‘11 and were poised for a big bounce back the next year (Bull Market). My Stock Market Indicator proved to be spot on as the Buckeyes were the Most Improved Team in the country going 12-0 in ’12! In ‘13 Auburn was coming off a 3-9 season following two years of a combined 22-5. That earned them a +8 in my Stock Market Indicator and they got to the National Title game! In 2014 TCU was +5 in the Bull Market and went from 4-8 to 12-1 and was the Most Improved Team in the country! In 2016 UCF was an amazing +10.5 in the Bull Market and improved from 0-12 to earning a bowl bid. Last year 7 of the 9 teams that made the Bull Market box in ‘17 went from a losing record to a bowl!
The Bear Market has also been a very accurate indicator. Of the 8 teams that made 2016’s article, only 1 managed to improve their record with 6 of the 8 having weaker records. Last year 9 teams were in the Bear Market box including Colorado, Idaho, Tulsa, Eastern Michigan, Western Michigan who all missed out on a bowl after getting to one in 2016. Here is a breakdown of the records and this years qualifiers.

THE BULL MARKET: After several years I expanded my research to take a look back at all records since 1990 and the numbers were proven to be solid for decades. Since 1990 there have been a total of 30 teams that had a +6 Stock Market Indicator meaning that the prior year’s win total was 6 wins less than the average of the 2 years prior. Of those 30 teams, only 3 teams have had a weaker record, while 26 improved and 1 stayed the same. That improvement was dramatic going from a combined 84-272 (23.6%) to 166-199 (45.4%). Teams in the +5.5 category have improved their record 17 times had a weaker record just 3 times and the same record twice. There have been 43 instances of teams being +5 in the SMI category and those teams were 32-10-1 (76.2%). The records for +4.5 are now 44-13-3 (77.2%) and despite going down to +4 in my Stock Market Indicator, it surprisingly jumped back up to 53-13-6 (80.3%), which are superior to the individual +4.5 and +5 range. The numbers do drop back to 73.4% for a +3.5 net wins, 72.9% for +3 and down to 64.6% for +2.5. However, just when you thought the trend would get lower, surprisingly +2.0 has a 71.0% success rate. Even teams that had a +1.5 SMI from the prior two years, went up or stayed the same 66.1% of the time and teams that were either +1 & +0.5 combined for 64.9%!

Summing it all up, I will put it into 4 categories. Teams with a +6.0 SMI or higher are an almost perfect 26-3-1 (90.0%). Teams in the +4 to +5.5 range are 146-39-12 (78.9%) while teams in the +2 to +3.5 two-year SMI range are 272-131-43 (67.5%). Teams that were in the +.5 to +1.5 range are 331-212-67 (61%). Overall, of the 1,283 teams s/’90 that had a +.5 SMI or higher, 775 improved their record, and 123 had the same record meaning 70% of the time, teams in that category improved or had the same record.

This year 4 teams fall into the top bracket of +6 or more on the two-year SMI. Also listed are teams with a +5 or more on my SMI.

 

THE BEAR MARKET: Like all of my systems I put in my magazine, there is a two-way street to this Stock Market Indicator (SMI). I also have reviewed the numbers since 1990, on the Bear side of the market, teams with a negative indicator, indicating that a team had likely over achieved the prior season, and the numbers were similar to the Bull side. In fact of 43 teams that had a -6.0 SMI or lower since 1990, only 3 managed to improve their record. Troy was one of those rare exceptions last year going from 10-3 to 11-2. The biggest drop-offs came in 1998. Tulane had a -7.5 SMI and went from 12-0 to 3-8. Interestingly in 1998 I called for Washington St to go from the Rose Bowl to last place in the Pac-10, which they did and in the 2 year SMI they were -6.0 and went from 10-2 to just 3-8. Last year Tulsa was -6.0 and went from 10-3 to 2-10.

A look at the numbers shows a powerful trend of basically 77.3% if teams have a -2.0 SMI or less since 1990 have had a weaker or the same record the next year. Unlike the above positive Market Indicator, there were actually two teams that bucked the trend with a -7.5 or -7 and improved but overall the chart has higher percentages at the -6 & below level (37-3-3 92.5%). Teams that had a Stock Market Indicator of -5 or lower had a weaker or the same record the next year 90% of the time with 102 having a weaker record, only 12 with a stronger record and 8 having the same record. Even teams with a SMI between -3.0 to -4.5 were in a 75.9% category. The teams in the -2.0 to -2.5 Stock Market Indicator had a weaker or same record the next year 73.1% of the time. Teams in the -1 to -1.5 range had the same or weaker record only 63.6% of the time. Overall, teams that had -2 SMI or lower had the weaker or the same record basically 77.2% of the time.

 

The last two years 18 years rated a -5 or higher (in the box of this article), and only 5 improved their record. Here are the 6 teams with a -5.0 SMI or lower  including 5 at -5.5 or higher.

Here is the complete list of the Stock Market Indicator. Teams highlighted in green are the Bull Market. Teams highlighted in red are the Bear Market.

 

PHIL STEELE PLUS

Get Phil’s Extended data and Plus your odds. Click on Learn more to see exactly how Phil uses Phil Steele Plus. FCS only access also available.