You had the right idea - and almost the right implementation. The function "order_target_percent" makes it's calculations based on the number of filled shares, and ignores any open orders to reach the target amount. To avoid the wild swings back-and-forth, you should wait to rebalance until there are no more open orders. So the portfolio weights are maintained. I cloned your algo and added "if get_open_order():return" on line 6

Thank you for the thoughtful response. I've many more questions, but have only a few hours here, so I'll dig deeper before papering the community with questions. Understanding Excel logic functions and Google Scripts, I thought this would be a little easier, but can't seem to find a cohesive and comprehensive library of available code logic and argument building blocks beyond the API reference page. If you have any pointers, they would be welcome.

A few issues remain that I will throw out there. I really am searching for the underlying key to unlocking each question for myself as it will make me go much faster. If anyone can point me in a direction or has the time to respond, that would be a fantastic start to what I am trying to accomplish...

1. Day 1 does not execute an order which fills the allocation - why?
   
2. If I wanted to assign a separate allocation to cash, what would be proper form?
   
3. Can cash go negative? If so, how do you prevent margin?
   
4. Can a drift collar be assigned to the effect of +/- 1% of allocation after initial allocation fulfillment...including a separate cash allocation?
   
5. Is there the equivalent of a MIN and/or MAX feature? Arrays? IF/AND/OR? Date function and Date arguments: IF(Date X>=Date Y?, Date X - Date Y, TODAY) ....I'm sure there are very eloquent and specific ways that can be accomplished here...I just have to find the mental key.
   
6. How are ETF orders rounded to an integer?
   
7. If dealing with several tickers and arguments, how do I circumvent the apparent limit of 5 arguments for "record"? I may have a limited understanding of the very question I am asking here.
   

To answer your questions,

1. In the backtester, orders are submitted in one bar and filled in the next bar. They will be filled according to the slippage model. So in daily mode (which is what you had) an order submitted on Monday will get filled on Tuesday at the end of the day. In minute mode backtesting, an order submitted at 9:31AM will get filled at 9:32AM. I would recommend to backtest in minute mode because you get more frequent price data and it's closer to simulating live trading. Just because your algo runs every minute, that doesn't mean it has to trade that frequently. You can use schedule_function to specify the day and time to trade: https://www.quantopian.com/help#api-schedulefunction

2. I'm not sure what you mean, could you clarify? Maybe with an example?

3. Yes, cash can go negative. If you want to prevent this, you can add it as part of your code and here's an example. Also, the "order_target" family of functions will seek to your target positions and should avoid dipping into negative cash if used properly. In case you haven't seen them, you can also set trading guards in your algo.

4. Are you looking to prevent drift by rebalancing? If so, the order_target functions will keep your target amounts as your portfolio shrinks and grows. Otherwise, could you share an example (in words) of what you're trying to do?

5. You can do direct date comparisons, here is an example: https://www.quantopian.com/posts/rebalance-algo-9-sector-etfs. However, I'd recommend to use schedule_function instead, it's simpler to code and much easier to apply the datetime comparison.

6. Orders are rounded down to the nearest whole share.

7. record() is limited to 5 parameters, but if you want to plot more custom variables then try the research environment. It's currently in beta, and if you don't have access, you can jump the waitlist by submitting an entry to the contest.

Let me know if that helps and if you have any other questions!

Alisa