Ah yes. The missing semicolon was the problem. Thanks!

@drew, I got the error "incorrect syntax near the keyword 'with'.
WITH Pinhead_rn AS
(
  SELECT *, ROW_NUMBER() OVER(PARTITION BY [ID] ORDER BY checkMe DESC)...

Awesome. Thank you!
The first 2 lines
DELETE FROM A
FROM dbo.PinheadTest a
I didn't imagine that could be done like that.
DELETE FROM someTable FROM someTable...

Awesome. Exactly the kind of specific guidance I needed!

Thanks, Thom. Very clear description.

John Mitchell-245523 (12/6/2016)

---

Not sure whether this is an option for you, but my suggestion is just to ditch the loop. You can replace the whole lot with something like...

Uh... no. I assume the while...

Excellent write-up! Thanks again.

So if I wanted to adjust the code and expand on the portion that builds the random value added to the base value, I could just use...

@Eirikur, Much thanks! The output is nearly there. Now I need to understand it.

I got as far as

(SELECT N FROM (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) AS X(N))

I figure you chose 10 values arbitrarily...

Thanks for responding.

I knew of TOP. But I didn't want one row. Which is why I wrote

.

Anyways the SSRS dialogs for setting that...

I will forward that suggestion on. Thanks for the info!

The error message from the tooltip over either of the red-underlined column names is: "... is of a type that is invalid for use as a key column in a...

This is the query I used. I extrapolated because I thought I understand the model your query involved.

Basically the index must cover each of the fields being compared on.

So table...

ScottPletcher (8/4/2016)

---

Given that there is no clustered index now, you might as well clus to match this query, something like below. If you have DATA_COMPRESSION available, be sure to...

Thanks for the clarification. Luis pretty much said that but I wanted to make sure..

Luis Cazares (8/4/2016)

---

With 1.2 and 427 million rows, you really should add an index to those tables. You should start by defining the clustered index and add non-clustered indexes as...