Reporting Transactions

[Demreb]

Another topic that came up during the season and was discussed at The Banquet was how to report weekly transactions.

Currently we report them to David via e-mail or voicemail and he logs those moves at the TQS website.

TQS has the capability of allowing each owner to record their own moves and it's been suggested that we have TQS "turn on" this feature (which doesn't cost any additional money) for the CFCL.

There are pros and cons to doing this, but before we get too detailed on that, what are your thoughts?

[Copperfields]

Some input from Teddy:
I've already mentioned this before, but the current manner of submitting moves and bids is extremely inefficient with little to no control for errors. I understand that some owners may live in the dark ages with respect to computers, but I think it is less than optimal to bring the technology and capability of the league as a whole down to their level. I would suggest at a minimum that there at least be an option, not necessarily a requirement, to take advantage of the on-line tools at TQS. I understand that the TQS tools are not as powerful as the complicated IF-THEN nested transactions that are currently possible, but it seems that it would be practical to at least give owners the use the on-line tools as an option.

[stones]

I am a big fan of technology. I would be supportive of allowing owners to make their own moves on TQS or other service. I am currently in a couple of H2H Football Leagues (on CBS Sportsline) that allow owners to make their own moves.

BUT... I would never want to minimize the value of David's passion and contribution to and for this league (and Rich's when David takes that very infrequent "time off"). He has done an amazing job keeping everything moving smoothly throughout the season and in particular on 'special days" like AL free agent day or trade deadline day, but also let's not minimize Friday afternoons and evenings with the transactions and weekly report posting.

While I am sure there would need to be some adjustments needed to make sure we all have the same opportunity to review trades involving free agents from AL or positional eligibility issues, for the most part I think it would be worth the effort.

I will admit that I am not familiar with TQS and its process, so I will leave that up to those who have actually worked with it in other leagues.

I would also hope that by embracing technology we don't lose the weekly newsletters/report postings. In fact, I think by David's own admission, those have gotten shorter and shorter over the course of the past few seasons, I think due primarily to his family commitments (which are always the absolute most important commitments). If we cut down considerably on the time required for the Commissioner to make appropriate moves, maybe David or others will be re-inspired to prepare more detailed and substantive weekly reports. Those are always a highlight of my week.

Just some random thoughts from a random guy!

[Copperfields]

Steve,
Thanks for the comments about the work Rich and I put in to get the transactions entered, and also about the weekly transaction report -- I'm never sure who (if anyone) reads those, so it's great to hear them referred to as a "highlight" of someone's week -- though that may say more about your week than the reports!

Since I'm a little closer to the whole transaction deadline/processing topics than most owners, I want to wait until others share their opinions before chiming in with mine. However, one issue I did want to raise is the topic of whether allowing everyone to submit their own moves at TQS would cut down on the time committment from whomever is filling the League Secretary's role in a given week - there's a chance that such a chance would actually create more work for the Secretary. I don't know that for sure, just raising the possibility. For example:

• Not all owners have regular access to the Internet (or their access to TQS is blocked, or they simply are not comfortable working with the web form for entering moves), so chance are good that some owners would enter moves at TQS, others would send them via e-mail, and still others via voice mail. That would mean that the Secretary would need to gather transactions from three sources to resolve bids and waiver claims, and then potentially have to back-out winning bids and related moves at TQS. Moving the transaction date to the weekend could potentially help, since most everyone would have Internet access (though, some, like Matt G., still might not, depending on his schedule at the firehouse).
• The moves entered at TQS would still need to be reviewed -- my experience with TQS indicates that they do not check transactions for the some problems, such as: salary cap violations, positional requirements, roster sizes (ie, won't remind you to waive a player when you acquire a free agent). These are things I currently to for all roster moves (I update the Excel rosters, which includes checks for all of these things before I enter moves at TQS). This could lead to some messy situations with having to back out moves and re-enter the correct ones.
• We have a number of special restrictions on when and how certain moves can be made that TQS will not be able to track: minor leaguers who cannot be activated before a certain date, asterisk players who cannot be waived or released, the rules about when mid-week moves are and are not allowed, etc. These are all things that would need to be checked and possibly backed-out.
• Things like salary changes (for waived players) and contract changes (for minor leaguers promoted) are not automatically done by TQS, so individual owners would need to do it manually

These are just some of the things that could potentially cause additional work for the Secretary. I need to emphasize that I'm not necessarily against allowing owner-input of moves -- there may be ways to address all these concerns. I just wanted to point out that making this change might make more work for the Secretary.

I'd love to see additional discussion on this.

David

[redhots]

I also am a big fan of technology and would love to see us be able to enter our own roster moves at TQ. The CFCL does have special rules that would need to be looked at requiring special considerations but I think overall it would benefit the majority.

I too love the weekly transaction report and while I do not share the "highlight of my week" sentiment of Steve, it is something I look forward to reading and enjoy.

[MGrage]

My father was a Luddite and my mother smelled of elderberries so you can guess how big a fan of technology I am.

Not really, but I'd still rather stick with the status quo. I think that old cliche applies, if it's not broken, don't fix it. I believe that making this change would create a whole new set of problems and that on the whole, it wouldn't be worth it. Mahalo

Matt

[Splinters]

mgrage said:

I'd still rather stick with the status quo. I think that old cliche applies, if it's not broken, don't fix it.

I disagree - I think that the system is broken.

Once this season, I placed the highest bid on a free agent but the bid was not able to be processed because I did not make the necessary secondary moves to enable my starting roster to fit within the salary cap. I completely blame myself for the mistake; however, the only error checking system that we have in place is the need to MANUALLY verify (1) position qualifications, (2) size of roster, (3) total salary of starting lineup (my problem), etc.

A better system would take advantage of AUTOMATED error checking.

[MGrage]

Well, no system is perfect, even an automated system. Power outages, high Internet traffic and slow connections can even prevent you from making bids and/or changes. Plus, I like the orderly nature of the way we do things now. Plus, if I understand Dave correctly, TQS does very little error checking if at all.

The only other fantasy league I'm in right now is a football league over on CBS Sportsline. There you can activate/reserve players at will until an hour before the first game of the week. I have no idea how that works for baseball with games every day and at differing times. There are no salaries and if you want to pick up a FA, you send in a post to the league message board. You can then add/drop the players or have the Commish do it.

I guess we'll just have to disagree that the procedure is broken. I think it works 99% of the time due mainly to Dave's diligence. I like the flexibility of the current system too. I dunno IF there is a program that can do all the tasks that Dave has to do each week, that might be OK. But that's a big if and I'm not sure that even exists. That would save Dave alot of time each week so that's a good thing. Then again, we might lose Matt because he doesn't want to use the Internet. Mahalo

Matt

[Splinters]

mgrage said:

Then again, we might lose Matt because he doesn't want to use the Internet.

Take a look at the initial proposal - "an option, not necessarily a requirement, to take advantage of the on-line tools"

[Demreb]

I agree with Matt on this one. I don't believe the system is broke. While I can appreciate the frustration Teddy had when he lost out on Hillenbrand, the reason he lost out on him was he submitted an incomplete bid.

The Constitution clearly lays out that if a bid is incomplete (or a trade and subsequent moves is incomplete) then the whole thing becomes null and void.

Now David has been a nice guy many, many times in the past to try and have what an owner wants actually happen. BUT it is our responsibility to give him the complete information from the beginning. Sometimes he has to make a call (like disallowing the Hillenbrand move) because he couldn't stretch the rules without making it unfair to the league as a whole.

I don't think that us being able to enter our transactions at TQS last year would have prevented the same mistake. While I was covering for David last year I ended up screwing up a few transactions, had to "undo" the transactions and then redo them. The only reason I knew I screwed up the transactions was that I knew what to look for. TQS didn't prompt me by saying what I wanted to do was wrong.

What I know about the reporting, I think this would dramatically increase David's time commitment (which isn't fair to him) and slow the processing of information to us.

But I will lay this one on the feet of David. If he wants to take this on, that's fine. But overall I'm against it.

[Copperfields]

splinters said:

Once this season, I placed the highest bid on a free agent but the bid was not able to be processed because I did not make the necessary secondary moves to enable my starting roster to fit within the salary cap.

Just to clarify, the free agent bid was not nullified because of salary cap restrictions, but because a player wasn't waived to make room for the new guy on the 40-man roster.

splinters said:

I completely blame myself for the mistake; however, the only error checking system that we have in place is the need to MANUALLY verify (1) position qualifications, (2) size of roster, (3) total salary of starting lineup (my problem), etc.

Submitting moves directly to TQS would do little to reduce the amount of manual verification in these cases. Owners would still need to manually verify items in all three:

1) TQS will flag an error if you try to put Albert Pujols at SS, but it won't stop you from having nine OFs active at once.

2) TQS will flag an error if you have 24 active players, but not if you've got 23 active and 19 on reserve (in other words, TQS would not have flagged an error in the case of your free agent bid described above -- it would not have forced you to waive a player, which is what nullified the bid. That check would still need to be made manually, so submitting moves to TQS would not have made a difference).

3) I do not believe TQS performs any check on salary caps. I check total salaries in the Excel spreadsheet before inputting moves to TQS, so I usually catch violations before it ever gets to the TQS stage. However, I do not tell TQS when our salary cap increases at various points during the season, and it never flags errors when teams go over $3.00 after the All-Star Break, which leads me to believe it won't flag errors at any point.

In other words, TQS would provide some level of error checking on roster moves, but there would still need to be significant manual verification by the owner prior to submission -- or worse, after submission by whomever's performing the League Secretary's role. I say "worse" because it would mean significant work for the League Secretary to verify and possibly back-out moves that had been incorrectly submitted to TQS.

All that said, I'm perfectly willing to explore a situation where owners have the option of submitting their moves on their own, as long as there were certain requirements and caveats in place (things like which kind of moves can/cannot be made by owners, what happens if errors are discovered, etc.). I also need to look more closely at how the whole process works. If we can set up a process where moves are submitted to TQS but aren't made effective until approved by the Secretary, it might reduce/eliminate some of the rework I'm concerned about.

Unfortunately, TQS does not have a good sample league available for playing around with, and they've already disabled our ability to make changes to our league set-up.

I do know that there are vast numbers of leagues out there who allow users to submit their own moves and the majority of them love the functionality. I do not know, however, if they have the depth and complexity of rules that we have to contend with.

David

[Copperfields]

I think it's worthwhile to re-open this discussion and maybe try to gather some information.

Part of the problem I have in evaluating the possibilities of owners entering their own transactions at TQS is that I've never done it (as an owner as opposed to Secretary) and haven't seen the interface. TQS has shut down the ability to make changes to our league (presumably until we re-up for 2007), so I can't turn the feature on to see how it works. As such, I'm reliant on those who have used the feature in other leagues to describe how it works.

One main question I have: If you enter your own moves at TQS, do those moves go through right away, or do they have to be approved by the Commissioner before they become active at TQS?

If we're going to have some kind of hybrid model, where some owners enter their own moves and others submit them to me via the traditional approach, will I have the chance to look at the moves that were made at TQS (to sort out bids, waiver claims, etc) before those moves go into effect at TQS?

For example, say Steve enters a bid at TQS for free agent Don Money and his is the highest of all bids sent to TQS for Money. However, say Rich had sent his moves via e-mail and those moves included a bid for Money that tops Steves.

Will I need to go into TQS, remove Money from Steve's roster, undo all the associated moves, reimburse his FAAB balance, then assign Money to Rich's roster, etc.?

[Nick's Picts (archived)]

I am all for e-automation whenever possible. It's job security for me.

Be that as it may, however, I wanted to add my experiences with developing software specific to this league and our rules. Basically, I've been toying with a draft and league manager application for the last several years with an eye on open sourcing the code base. That being the case, I have been working from standard fantasy setups and then adding in the bits that would be required to use the applications in our league. On the one hand it forces me to write good, modular code. On the other, it is a royal pain in the arse to account for all the idiosyncrasies of the CFCL.

The point of this is that I would hesitate to rely on what passes for error checking for TQS or any other service can provide as an automated service with more accuracy than the current system. We fall under the 80 - 20 rule* and no one without ulterior motivation is going to have written software that accommodates constraints like shifting salary caps, qualifications for position eligibility, contingent transactions, and dynamic priority for free agent bids based on rankings. Heck, I find I have to scrape various web sites just to get accurate Holds+Saves data to run projections on. Volume stats services cannot be depended upon to provide accurate confirmation of league transactions. This would seem especially true for TQS as David reports they have not implemented some fairly basic and widely applicable exception checking.

At any rate, I would argue we keep the system as it currently exists unless someone finds a reasonably priced service that can accommodate our particular quirks or I finally finish my freaking project and make it available to the league. As for whether the system is broken or not I would argue it is not. The system, while not entirely idiot proof**, functions as advertised.

*In software development 80% of your effort is expended trying to reign in the last 20% of your requirements. In other words, you can please most of the people some of the time and unless you've got bottomless pockets, that's going to have to be good enough.

**No offense meant. It's a term of art in the field.

[Splinters]

nickpicts said:

The system, while not entirely idiot proof**, functions as advertised.

**No offense meant. It's a term of art in the field.

No offense taken. I fully admit to being an idiot. This is the reason an automated error-minimization system makes sense to me.

[Nick's Picts (archived)]

splinters said:

This is the reason an automated error-minimization system makes sense to me.

My point being, it is extremely unlikely that an automated error-minimization system is likely to exist without you or someone else sufficiently familiar with our rules becoming motivated to put together an algorithm that accounts for the myriad details that make transactions in the CFCL so complex. Automated systems are good at computation and batch processing, not so good at applying logic toward--and even worse at interpreting--a data set.

In the case before us, this weakness is further complicated by the many different data stores. Position eligibility lives on TQS, salary cap information lives in the rules and is determined by date which is another data store, roster limits and positional requirements also live in the constitution, &c. An automated error-minimization system is going to have to be able to interface with all of these data stores, have an accurate algorithm for processing individual transactions based on the rule set, and manage the contingencies that come in to play when multiple teams are making multiple transactions simultaneously with the additional requirement that these transactions hold internal contingencies themselves.

Speaking with experience, this kind of recursive programming is a tight kitty. As such a system is not likely to exist, the present system is by far the most accurate one in existence even with all of its inherent flaws. If, on the other hand, I am wrong and such a system does exist, then by all means I vote we adopt it.

At any rate, I imagine this post comes off with a high stink of pedantry. For this I apologize. However I wanted to underscore my previous point because I did not make it explicit and your response indicated that perhaps I should have elucidated further.