These results are based on applying my algorithm as-is, without modification or adjustment of any kind, to actual NYSE/NASDAQ data over one year, ending about a month ago.

After reviewing this data, if you would like to have access to the Members Only area, which shows the daily output of my algorithm, please click here to sign up.

The detailed data below shows the actual, unedited performance of my algorithm.

Yuyostox: Advanced AI Technical Analysis - Results Over One Year
From 2018-09-23 to 2019-09-23

Important: See notes below table.
These results are for up to a month ago.
For up to date data and daily updates, please sign up as a member.


Buys: 76
Profits (avg): 50 (6.33%)
Losses (avg): 26 (-5.09%)
Success Ratio (# of profits/# of trades): 66%
Avg calendar days held: 12.4

Sell DateStockSell PriceBuy DateBuy PriceSignalDays
2019-09-13KEM19.89 (9.89%)2019-09-1018.1018.353
2019-09-06SBGL5.08 (-6.96%)2019-08-285.465.489
2019-09-05GFI5.63 (-7.10%)2019-08-266.066.1010
2019-09-04EGO9.91 (7.14%)2019-09-049.259.260
2019-08-30LDOS87.59 (7.17%)2019-08-0281.7383.4128
2019-08-23HLT91.91 (-2.03%)2019-07-2493.8195.1230
2019-08-19ARMK40.91 (9.21%)2019-08-1237.4638.107
2019-08-19ARWR33.38 (5.73%)2019-08-1931.5731.190
2019-08-15JBLU18.15 (-5.81%)2019-07-1719.2719.5029
2019-08-15GOL18.08 (-2.27%)2019-08-1418.5019.271
2019-08-08AIG57.29 (6.80%)2019-08-0753.6454.381
2019-08-07TGNA14.36 (-6.27%)2019-07-3115.3215.497
2019-08-05BX44.69 (-6.19%)2019-07-2347.6447.8413
2019-08-05TSS125.26 (-6.34%)2019-08-01133.74135.724
2019-08-02IOVA23.15 (-5.55%)2019-07-2924.5124.904
2019-08-01ATUS27.42 (7.24%)2019-07-1825.5725.8014
2019-07-31IFF144.87 (0.78%)2019-07-01143.75145.0930
2019-07-26TEAM145.96 (9.64%)2019-07-25133.13135.351
2019-07-25CPB40.35 (-0.49%)2019-06-2640.5541.0429
2019-07-24TER53.61 (16.44%)2019-07-1946.0446.305
2019-07-23XRAY56.36 (-3.19%)2019-06-2458.2258.8829
2019-07-19KBR26.05 (6.67%)2019-06-2124.4224.5628
2019-07-16PNC139.13 (2.82%)2019-06-17135.31135.2129
2019-07-16AAL34.63 (5.32%)2019-07-1132.8832.945
2019-07-15DAL62.37 (5.77%)2019-07-0958.9759.376
2019-07-09MOMO33.84 (-7.34%)2019-07-0236.5236.737
2019-07-03OKTA129.57 (6.47%)2019-06-28121.70122.745
2019-07-02DISH41.06 (6.43%)2019-06-1838.5838.9914
2019-07-02SNPS132.83 (5.42%)2019-06-19126.00125.8613
2019-07-01EDU100.93 (6.15%)2019-06-2795.0895.624
2019-06-21KMX87.56 (7.50%)2019-06-1181.4582.3610
2019-06-19MTG13.82 (-0.79%)2019-05-2113.9313.9429
2019-06-18CMCSA43.58 (5.52%)2019-06-1041.3041.488
2019-06-18SPWR9.95 (20.46%)2019-06-148.268.464
2019-06-14KBH26.42 (-1.20%)2019-05-1526.7426.8930
2019-06-11PYPL117.41 (6.21%)2019-05-29110.55111.7913
2019-06-10EVRI12.11 (6.32%)2019-06-0411.3911.416
2019-06-10TNDM67.14 (-4.68%)2019-06-0570.4471.125
2019-05-24ARQL7.73 (9.65%)2019-05-227.057.172
2019-05-14KBH27.10 (6.27%)2019-04-2425.5025.1220
2019-05-10BLL63.82 (6.21%)2019-05-0360.0960.007
2019-05-07HPE15.48 (-6.86%)2019-04-2216.6216.6615
2019-05-07STM17.17 (-6.74%)2019-04-2518.4118.6112
2019-05-03SAND5.19 (-4.95%)2019-04-055.465.4928
2019-05-03TPH13.76 (5.68%)2019-05-0113.0213.052
2019-04-29TAL38.56 (6.76%)2019-04-2636.1236.173
2019-04-25WCN92.88 (5.71%)2019-04-0487.8688.2521
2019-04-24XLNX139.88 (6.69%)2019-04-11131.11131.5913
2019-04-22HBM7.04 (-5.38%)2019-04-157.447.657
2019-04-18MRVL25.23 (3.61%)2019-04-1724.3524.011
2019-04-17ARRY22.96 (-5.05%)2019-04-0824.1824.449
2019-04-10INCY83.36 (-0.67%)2019-03-1283.9284.1729
2019-04-05GLUU11.66 (5.81%)2019-03-2611.0211.0210
2019-04-02PTLA36.92 (5.60%)2019-04-0134.9634.701
2019-03-21WPM24.00 (6.69%)2019-03-1822.5022.583
2019-03-18DXCM154.86 (6.09%)2019-03-14145.97147.464
2019-03-12AEM44.43 (4.00%)2019-03-0842.7242.294
2019-03-04AZUL28.91 (-7.25%)2019-02-0531.1730.9727
2019-03-04NBEV5.63 (-3.35%)2019-03-015.825.873
2019-02-27TSN62.58 (1.26%)2019-01-2961.8061.8029
2019-02-25DECK149.09 (6.45%)2019-02-07140.06141.3118
2019-02-22CNP30.97 (3.68%)2019-01-2429.8729.8729
2019-02-15XRX30.46 (5.12%)2019-02-0628.9828.989
2019-02-14INCY83.97 (5.03%)2019-01-2879.9579.9517
2019-02-05ENIA10.44 (5.17%)2019-01-149.939.9322
2019-02-04CDNS49.98 (7.21%)2019-01-2246.6246.6213
2019-02-04MEDP67.42 (5.50%)2019-01-3163.9063.904
2019-01-25DUK86.31 (0.41%)2018-12-2885.9685.9628
2019-01-25RMD97.41 (-15.59%)2019-01-11115.39115.3914
2019-01-25HRC103.02 (7.89%)2019-01-1795.4995.498
2019-01-15INCY79.10 (5.61%)2019-01-0974.9074.906
2019-01-08EXEL23.22 (6.02%)2019-01-0821.9021.900
2019-01-02SCG47.70 (-3.83%)2018-12-2149.6049.6012
2018-12-06ESRX97.57 (0.48%)2018-11-0697.1097.4530
2018-10-25DXCM123.36 (-6.49%)2018-10-16131.91133.209
2018-10-11MUR36.11 (6.82%)2018-10-0533.8133.996
Green: profit, Red: loss, Gray: liquidated after 29 days.


The figures above are based on the buy signals output by my algorithm, and show the results of simulated transactions based on actual stock data. Analysis is based on my data feed of actual stock prices from CSI Data, all results E. & O.E.

Buy dates shown are the next trading day after a buy signal. Buy prices are the average of the low and the signal price, or the low itself, if the low exceeds the signal price. Sell prices are triggered at 7.5% down or 5% up from the signal price, or after 29 days.

When triggered, actual sell price is avg(hi+low) for stop loss or 29-day selloff, and avg(hi+sigprice) for profit.
Dividends are not reflected in ROI calculations. For convenience, I force liquidation of holdings upon stock splits, conversions, etc.

I developed this algorithm myself, and distribute its daily output as a paid service. I do not receive a penny from any companies, stock promoters or anyone else to recommend any given stock. I may trade according to my own algorithm's output, but I only purchase after the output is published, and the algorithm is never adjusted to suit my own holdings, or in any other way.


-Jul 2017 (32%)
-Aug 2017 (34%)
-Sep 2017 (49%)
-Oct 2017 (49%)
-Nov 2017 (91%)
-Dec 2017 (68%)
-Jan 2018 (81%)
-Feb 2018 (96%)
-Mar 2018 (92%)
-Apr 2018 (102%)
-May 2018 (109%)
-Jun 2018 (130%)

The following notes are important in order to properly understand the results.


These figures are the results of applying my A.I. algorithm to actual stock market data. You will see that the recent results as well as the earlier one-year periods above show profits significantly superior to, say, the S&P 500 index.

Regarding the previous results up to Jun 2018, every one of the twelve one-year periods resulted in a profit of 30% to 130%! And 10 out of the 12 periods yielded annual profits of at least 49%! All this, simply by running the algorithm with no tweaking or parameterization.


The following are the assumptions upon which the twelve consecutive one-year tests were based. Note that my algorithm generates signals based on its artificial intelligence assessment of actual stock market data, without taking into account these twelve tests at all.

The test method is simple: take the signal and simulate a stock market buy based on the indicated price, if available. Note that the signal is based on the previous day's closing price. If the current day's price has moved more than 1% above the signal price, and it is no longer possible to purchase at or below the signal price, the signal is not acted upon. Interestingly, this almost never happens.

Brokerage and taxes are not taken into account. Proceeds are compounded. For the test, if a signal arose while a position was being held, there was no cash to act on the signal, so it was ignored.

  • Since prices are always fluctuating, it is not possible to state precisely what the buy or sell price is, so for automated testing purposes, when a buy signal appears, we assume that the actual buy price is the average of the low of the current day and the buy signal (which is the close of the previous day).
  • Similarly, when a profit target is reached, we assume that the sell price is the average of the high for the day and the profit target, and when a stop loss limit is reached, we assume that the sell price is the average of the high and the low for the day.
  • The above approximation of trading prices is in our view a reasonable and realistic way to perform completely automated testing. Other than the start date, there was absolutely zero human input for each one-year period, ensuring the removal of any subjectivity.

    Technical details

    Yuyostox is not a statistical model, but a factor-based predictive model incorporating artificial intelligence techniques. Rather than attempting to "fit the curve" by fiddling with the algorithm until it appears to work most of the time, which is hard enough to do but is generally unreliable, I developed Yuyostox from the start as a hypothetical model based on specific design principles. In particular, without referring to any statistical market data at all, I devised a way to measure intangible characteristics such as strength, comparative trading behavior (against industry peers), sustainability, etc. Nevertheless, I made it a point not to consider conventional stock fundamentals, such as details of the company, ownership, news announcements, etc. For full automation to be possible, I wanted to focus on pure technical analysis, based on price and volume data. In any case, I believe that fundamentals are fully reflected in the behavior of the market, as reflected in the technicals.

    Obviously, I was careful to ensure that the signals are generated with no lookahead of future data. Signals are decided based entirely on the A.I. looking at six months of data up to the closing price of the day previous to the signal date.

    I should also mention that my algorithm is based on a maximum of one month holding any stock (I sell after a month, regardless of profit or loss), a profit target of 5% per position and a stop loss limit of -7.5% within that window. On average, the profit/loss is realized well within two weeks (sometimes the very next day, as a 5% rise is all that is required). This allow for high turnover and compounding. It is definitely not buy and hold!

  • Sign Up