KOBA is a new proprietary model developed for the 2017-2018 season. As the successor of the retired ELF model, it uses many of the same algorithms. KOBA, however, has the edge on ELF and Vegas by leveraging state-of-the-art algorithms from active areas of research. While being computationally expensive, KOBA has been shown to provide the most consistent and profitable results.
The model was backtested using similar methodology as was used for ELF and RUS. Full Kelly Criterion was used to calculate the wagers.
The evaluation exercise undertaken here involves 8 seasons of past results (2009-2017) that are iterated through the following steps:
This process allows assessment of the model behavior with data it has not seen during its building phase and by having it run through all the games in a given season we get a feel for potential losing and winning streaks we might have to face.
Here are the model results from 2009-2017. There is no reinvestment of earnings. Note: the profit is in Units where 1% of bankroll = 1 Unit.
Compared to the two previous NHL models, ELF and RUS, KOBA shows stronger and more consistent profitability. Nevertheless, the returns are not linear and there will be ups and downs throughout any given season.
As always, we implore you to use this information smartly. Make sure to find a staking plan that suits your risk appetite and overall bank size. Do not bet beyond your own personal means.
All seven years previously modelled have shown to provide value, but the future is not guaranteed. We do not sell this service with the promise of profitability. We do sell this service with the promise of providing the same analysis the generated results detail above.