---------------------- Forwarded by Phillip K Allen/HOU/ECT on 06/12/2000 
10:55 AM ---------------------------
   
	
	
	From:  Tim Belden                           06/11/2000 07:26 PM
	

To: Phillip K Allen/HOU/ECT@ECT
cc:  
Subject: Thoughts on Presentation

It is a shame that the CAISO doesn't provide actual generation by unit.  The 
WSCC data, which is dicey and we don't have until July 1999, and the CEMMS, 
which comes on a delay, are ultimately our best sources.  For your purposes 
the CAISO may suffice.  I think that you probably know this already, but 
there can be a siginificant difference between "scheduled" and "actual" 
generation.  You are pulling "scheduled."  If someone doesn't schedule their 
generation and then generates, either instructed or uninstructed, then you 
will miss that.  You may also miss generation of the northern california 
munis such as smud who only schedule their net load to the caiso.  That is, 
if they have 1500 MW of load and 1200 MW of generation, they may simply 
schedule 300 MW of load and sc transfers or imports of 300 MW.  Having said 
all of that, it is probably close enough and better than your alternatives on 
the generation side.

On the load side I think that I would simply use CAISO actual load.  While 
they don't split out NP15 from SP15, I think that using the actual number is 
better than the "scheduled" number.  The utilities play lots of game on the 
load side, usually under-scheduling depending on price.

I think the presentation looks good.  It would be useful to share this with 
others up here once you have it finished.  I'd like to see how much gas 
demand goes up with each additional GW of gas-generated electricity, 
especially compared to all gas consumption.  I was surprised by how small the 
UEG consumption was compared to other uses.  If it is the largest marginal 
consumer then it is obviously a different story.

Let's talk.