I don't think you have understood what this graph shows, because it does account for your HODL strategy, let me explain.

First here's the numbers again for a single S9, starting up January 2018 (I've added a few more columns to make it clearer):
https://s13.postimg.org/jghsyuxp3/network_hashrate_prediction.jpg

It's all about the total network hashrate (and corresponding difficulty off course but they are dependent on eachother so I could have used either here, I chose hashrate).

The spreadsheet forecasts what the total network hash rate will do in 3 different scenarios:
- follow current trend (middle line)
- continue to grow faster and faster as it has always done in the past (top line, pessimistic outlook miners)
- stop growing and decrease again over the course of the year (bottom line, optimistic outlook for miners)

Based on this it predicts for the 3 different scenarios how much Bitcoin an S9 will make every month of 2018.
Up to this point this is all very solid data, you can pick your favorite scenario or anywhere in between, hashrate is not likely to go outside of these outer lines. You can get a really good idea of how much BTC your S9 will make every month over the coming year.

Now from this point onwards it becomes speculative as it depends on BTC/USD exchange rate, but that does not matter if you're trying to compare with a HODL scenario. I'll explain that later.
First I'll explain the rest of the spreadsheet:

The next columns show how much your miner will make you in USD based on today's echange rate, how much profit you get from that substracting power cost, and what your subsequent position is each month against your initial investment of your S9 (I've estimated $3000 to get an S9 incl. PSU into the country with shipping, tax etc.)

If you now look at the 3 position columns, only in the optimistic scenario (where hashrate stops increasing and eventually decreases again over the year) do you actually pay back your miner and end up in the green. Even following today's trend does not pay back your miner over the course of the year and by the end of the year you're only making $14/month so you never will either (I could add a few months to the spreadsheet to confirm this).

So to come back to the HODL comparison,
If yo put that $3000 into BTC today at $11500/BTC exchange rate, regardless of whether it goes up or down you will always remain in front compared to the current trend and pessimistic scenarios because the value of the initial investment fluctuates with the exchange rate, if you buy an S9 it doesn't (in fact it decreases because it depreciates but lets ignore that for now).

Now if you're claiming that you're not paying back your S9 investment with the BTC you earn and you're not paying the power cost with the BTC you earn but from another source of income, well, then in the other case where you just invest $3000 in BTC you should also convert that money from your other source of income to BTC every month and add that to the initial investment, so that investment grows equally from that other source of income. Otherwise you're not comparing apples with apples.

Again, only in the optimistic scenario where hashrate actually stops growing, is it economically sensible to invest in an S9 now at these prices and current hashrate growth, unless you have access to really cheap power.

I'd be happy to be proven wrong somewhere in my predictions/assumptions, but for now I can't see any way that would be the case.
Anyway, here's the adjusted spreadsheet again for those interested in playing with the numbers:
https://www.dropbox.com/s/v1m3xac87ptylj0/hash%20forecast.xlsx?dl=0