Planning to get started with XLB and XLE for now and add some more in the future.
Robert Lichello suggested along similar lines, but using stocks instead. Large mega-cap multi-nationals type stocks, household names that were born before you and I, and will likely still be around after we're gone. He also suggested equal weightings. Stocks held that are unlikely to go broke.
At a basic level consider 8 giants from across 8 different sectors and you equal capital weight $12,500 each, total $100,000 invested. For simplicity assume 'by coincidence' each stock is priced at $1/share at the time of purchase - as that aids in visualisation.
A year later, one has risen 20%, another fallen -20%, all others are unchanged. You rebalance back to equal capital weightings, which involves selling $2500 of the up stock at a price of $1.20/share so 2083 shares are sold to provide sale proceeds of $2500, and you buy 3125 shares of the $0.80 priced stock with that $2500. Overall the portfolio value is the same as at the start, all started with the exact same share price, and in total you held 100,000 shares. But after rebalancing you now hold 1042 more shares in total (1.04% more shares in total). If later all of the stocks all again realign to having a $1/share price, then buy and hold would be at break-even, whilst having traded (rebalanced) is up 1.04%.
If you accumulate dividends (cash) received during the year and reinvest those at the same time as rebalancing, then you in effect incur no additional cost to reinvest those dividends. You're simply taking the years best performer and selling some to add to the years worst performer, along with dividends also being invested.
Some investors suggest you should buy the entire market as most stocks underperform the 'average' due to a few stocks doing great. On the premise that unless you buy the entire haystack then you'll likely omit holding those few great stocks. But they're wrong. There's a fractal characteristic evident with stocks. Look at stocks, indexes, sectors, countries, or even smaller subsets and there's a common pattern evident, ranked from year worst to best and the left tail (worst) may be down a lot (but is limited to not being a greater than 100% loss), the right tail (best) may be up a lot, possibly even 100%, 200% ... or more. Rebalancing those two tails back towards near equal capital weightings can capture the effect I outlined earlier. Typically sectors/market/index/whatever left and right tails wont be the same pair of stocks being the worst/best each year - more typically they'll be different pairs. That said the last few years has seen Energy showing as a years bad performer in all three of the last three years, and Tech showing as the years best performer in two out of the last three years.
2017 +34.3% Tech, -0.9% Energy 2018 +6.3% Healthcare -18.2% Energy 2019 +49.9% Tech +11.7% Energy and Year to date 2020 Tech is again leading, Energy lagging.
Provided the individual stocks are sound/secure, inclined to survive no matter what, then generally it doesn't matter, your 'trading' (rebalancing) will average the average cost of stock down over time, as though you'd bought more stock at the start date than you actually did, which in turn also reflects through to the overall portfolio rewards achieved.
Lichello then took that "what to feed AIM" further, and added cost averaging using stock/cash variation over time (increase stock exposure after the market declines, reduce stock exposure after the market has risen at a above average rate).
Good luck with your sector ETF's, not that you'll need luck Many around here prefer ETF's for their greater resilience against going broke (cost-averager's nightmare). Yet others however are content with mega-cap multi-nationals. Lichello did also highlight how the Rockerfeller Trusts (back in the 1970's) comprised large amounts invested in relatively few stocks/diversification (predominately Exxon, IBM and Rockerfeller Centre Inc.). If you do hold individuals and even as few as 8 (12.5% weighting), a unlikely total loss of that 12.5% isn't that different to a broader market decline of -12% or more that at times occurs. Also, at times, individual stocks comprising a stock index such as the S&P500 can be weighted 10% within the Index, and/or individual sectors might be weighted 20% or more.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Many around here prefer ETF's for their greater resilience against going broke (cost-averager's nightmare). Yet others however are content with mega-cap multi-nationals. Lichello did also highlight how the Rockerfeller Trusts (back in the 1970's) comprised large amounts invested in relatively few stocks/diversification (predominately Exxon, IBM and Rockerfeller Centre Inc.). If you do hold individuals and even as few as 8 (12.5% weighting), a unlikely total loss of that 12.5% isn't that different to a broader market decline of -12% or more that at times occurs. Also, at times, individual stocks comprising a stock index such as the S&P500 can be weighted 10% within the Index, and/or individual sectors might be weighted 20% or more. 
Many around here prefer ETF's for their greater resilience against going broke (cost-averager's nightmare). Yet others however are content with mega-cap multi-nationals. Lichello did also highlight how the Rockerfeller Trusts (back in the 1970's) comprised large amounts invested in relatively few stocks/diversification (predominately Exxon, IBM and Rockerfeller Centre Inc.). If you do hold individuals and even as few as 8 (12.5% weighting), a unlikely total loss of that 12.5% isn't that different to a broader market decline of -12% or more that at times occurs. Also, at times, individual stocks comprising a stock index such as the S&P500 can be weighted 10% within the Index, and/or individual sectors might be weighted 20% or more. 
