What's everyone's chosen TSPCALC Strat for 2025?

[olijoel]

I've been using the seasonal approach with TSPCalc for three years, but really for two years blindly. Before that, I focused on the market, analyzing the candles and all that stressful stuff. In TSPCalc, I usually look for strategies with a mean around 20 or 25 and the lowest standard deviation. However, this year, I plan to adopt a different strategy that I'm still thinking about to determine if it makes sense. I intend to focus on identifying the strategy that provides the highest value when subtracting the standard deviation from the mean, regardless of whether the standard deviation is high or low. I do this because I’m considering the worst-case scenario for each strategy: which one would offer the highest return even if it reached the bottom end of its negative standard deviation.

[bloobs]

I've been using the seasonal approach with TSPCalc for three years, but really for two years blindly. Before that, I focused on the market, analyzing the candles and all that stressful stuff. In TSPCalc, I usually look for strategies with a mean around 20 or 25 and the lowest standard deviation. However, this year, I plan to adopt a different strategy that I'm still thinking about to determine if it makes sense. I intend to focus on identifying the strategy that provides the highest value when subtracting the standard deviation from the mean, regardless of whether the standard deviation is high or low. I do this because I’m considering the worst-case scenario for each strategy: which one would offer the highest return even if it reached the bottom end of its negative standard deviation.

Picking SD as a threshold low instead of the mean is a sound risk management rule. But please be aware that the SD value in TSPCALC (and MJ correct me if I'm wrong here) refers to the 1st SD from the mean. Some people may misconstrue this SD value as to refer to the 2nd SD.

In practical terms, if your chosen strategy has a mean of 25% and an (1st) SD of 5%, it means (no pun intended) this strategy is expected to return between 20% and 30% in any future year with ONLY A 68% CHANCE. Meaning there is also a 32% change it will return either less than 20% or more than 30%.

OTOH, if using 2nd SD, that 68% of a future chance of it falling between 20% and 30% chance goes up to 95%.

[tnwhiskey]

I've been using the seasonal approach with TSPCalc for three years, but really for two years blindly. Before that, I focused on the market, analyzing the candles and all that stressful stuff. In TSPCalc, I usually look for strategies with a mean around 20 or 25 and the lowest standard deviation. However, this year, I plan to adopt a different strategy that I'm still thinking about to determine if it makes sense. I intend to focus on identifying the strategy that provides the highest value when subtracting the standard deviation from the mean, regardless of whether the standard deviation is high or low. I do this because I’m considering the worst-case scenario for each strategy: which one would offer the highest return even if it reached the bottom end of its negative standard deviation.

Picking SD as a threshold low instead of the mean is a sound risk management rule. But please be aware that the SD value in TSPCALC (and MJ correct me if I'm wrong here) refers to the 1st SD from the mean. Some people may misconstrue this SD value as to refer to the 2nd SD.

In practical terms, if your chosen strategy has a mean of 25% and an (1st) SD of 5%, it means (no pun intended) this strategy is expected to return between 20% and 30% in any future year with ONLY A 68% CHANCE. Meaning there is also a 32% change it will return either less than 20% or more than 30%.

OTOH, if using 2nd SD, that 68% of a future chance of it falling between 20% and 30% chance goes up to 95%.

soooooo, is there anyway to utilize or find the 2nd dev in tspcalc?

[94aggie]

I am no statistics wizard, but this is great discussion. I came across this short video on youtube and it seems simple enough to add to the convo. Team, pls correct me if i am wrong in this example. Say a strat has a mean of 20% and 1 SD of 5%. Statistics would say that this strat has a 67% probability of a return of at least 15%.

Now to the 2 SD question. Again, the example above. 2 SD = 2*5%=10%. With a mean of 20% and 2SD means that this strat has a 97.5% probability of a return greater than 10%.

Now comes the ugly reality of the old disclaimer "past performance is no guarantee of future performance". Kinda like playing monday morning quarterback in a sense.

Check out this short video clip regarding 2 SD. https://www.youtube.com/watch?v=4spRfpkgcgQ

[bloobs]

I've been using the seasonal approach with TSPCalc for three years, but really for two years blindly. Before that, I focused on the market, analyzing the candles and all that stressful stuff. In TSPCalc, I usually look for strategies with a mean around 20 or 25 and the lowest standard deviation. However, this year, I plan to adopt a different strategy that I'm still thinking about to determine if it makes sense. I intend to focus on identifying the strategy that provides the highest value when subtracting the standard deviation from the mean, regardless of whether the standard deviation is high or low. I do this because I’m considering the worst-case scenario for each strategy: which one would offer the highest return even if it reached the bottom end of its negative standard deviation.

Picking SD as a threshold low instead of the mean is a sound risk management rule. But please be aware that the SD value in TSPCALC (and MJ correct me if I'm wrong here) refers to the 1st SD from the mean. Some people may misconstrue this SD value as to refer to the 2nd SD.

In practical terms, if your chosen strategy has a mean of 25% and an (1st) SD of 5%, it means (no pun intended) this strategy is expected to return between 20% and 30% in any future year with ONLY A 68% CHANCE. Meaning there is also a 32% change it will return either less than 20% or more than 30%.

OTOH, if using 2nd SD, that 68% of a future chance of it falling between 20% and 30% chance goes up to 95%.

soooooo, is there anyway to utilize or find the 2nd dev in tspcalc?

multiply SD by 2

[olijoel]

Picking SD as a threshold low instead of the mean is a sound risk management rule. But please be aware that the SD value in TSPCALC (and MJ correct me if I'm wrong here) refers to the 1st SD from the mean. Some people may misconstrue this SD value as to refer to the 2nd SD.

In practical terms, if your chosen strategy has a mean of 25% and an (1st) SD of 5%, it means (no pun intended) this strategy is expected to return between 20% and 30% in any future year with ONLY A 68% CHANCE. Meaning there is also a 32% change it will return either less than 20% or more than 30%.

OTOH, if using 2nd SD, that 68% of a future chance of it falling between 20% and 30% chance goes up to 95%.

soooooo, is there anyway to utilize or find the 2nd dev in tspcalc?

multiply SD by 2

Oh wow, yes. I understand. I will definitely take all this into account. Thanks

[tnwhiskey]

Picking SD as a threshold low instead of the mean is a sound risk management rule. But please be aware that the SD value in TSPCALC (and MJ correct me if I'm wrong here) refers to the 1st SD from the mean. Some people may misconstrue this SD value as to refer to the 2nd SD.

In practical terms, if your chosen strategy has a mean of 25% and an (1st) SD of 5%, it means (no pun intended) this strategy is expected to return between 20% and 30% in any future year with ONLY A 68% CHANCE. Meaning there is also a 32% change it will return either less than 20% or more than 30%.

OTOH, if using 2nd SD, that 68% of a future chance of it falling between 20% and 30% chance goes up to 95%.

soooooo, is there anyway to utilize or find the 2nd dev in tspcalc?

multiply SD by 2

thanks bloobs

[Bubba]

It would be interesting if MJ could implement a 2nd SD. I appreciate you mentioning that Aitrus. I hadn't considered that in my own perspective. Personally I use Sortino a lot, but that's more difficult to calculate with the current info.

Thanks!

Bubba

[12squared]

I've been using the seasonal approach with TSPCalc for three years, but really for two years blindly. Before that, I focused on the market, analyzing the candles and all that stressful stuff. In TSPCalc, I usually look for strategies with a mean around 20 or 25 and the lowest standard deviation. However, this year, I plan to adopt a different strategy that I'm still thinking about to determine if it makes sense. I intend to focus on identifying the strategy that provides the highest value when subtracting the standard deviation from the mean, regardless of whether the standard deviation is high or low. I do this because I’m considering the worst-case scenario for each strategy: which one would offer the highest return even if it reached the bottom end of its negative standard deviation.

Picking SD as a threshold low instead of the mean is a sound risk management rule. But please be aware that the SD value in TSPCALC (and MJ correct me if I'm wrong here) refers to the 1st SD from the mean. Some people may misconstrue this SD value as to refer to the 2nd SD.

In practical terms, if your chosen strategy has a mean of 25% and an (1st) SD of 5%, it means (no pun intended) this strategy is expected to return between 20% and 30% in any future year with ONLY A 68% CHANCE. Meaning there is also a 32% change it will return either less than 20% or more than 30%.

OTOH, if using 2nd SD, that 68% of a future chance of it falling between 20% and 30% chance goes up to 95%.

olijoel,
StdDev should be viewed as a measure of consistency rather than probability. Not coincidentally, the statistical term Variance is the square of standard deviation. Both can be used to calculate probability only if the distribution of the values is "normal", aka Gaussian.

bloobs,
I applaud your conviction to stay your course which put you underwater for several months while the S&P was up 10-20%.
Although you're currently +15% , it sure was a a bumpy ride.

12²