STILLWATER, Okla. -- Oklahoma State coach Mike Gundy said he wasnt worried when his Cowboys fell 17 points behind Iowa State in the third quarter.His explosive up-tempo offense showed why. James Washington caught eight passes for 152 yards and two touchdowns, and Oklahoma State beat Iowa State 38-31 on Saturday afternoon.The Cowboys trailed 31-14 with 5:55 left in the third quarter, but Washington put Oklahoma State ahead for good when he caught a 3-yard touchdown pass from Mason Rudolph with 3:40 remaining in the fourth. The 17-point comeback tied for the second largest in OSU history.We can score fast, Gundy said. If we were in a conventional offense, we would have been in trouble. Because we have the ability to score really fast, and we have playmakers who can pick up big chunks of yards, I didnt feel that way. But I didnt feel very good either.Rudolph passed for 351 yards and four touchdowns for Oklahoma State (4-2, 2-1 Big 12). Though the Cowboys gained 459 total yards, Gundy was disappointed in the teams offensive production. The Cowboys were held to 108 yards rushing, and most of that was gained in the final 20 minutes.We got a little bit impatient in the first half, Gundy said. Sometimes that happens. Even right now, I dont feel very good about the way we performed on offense. I thought we were very fortunate to come out of this with a win.It was the second consecutive late collapse for Iowa State (1-5, 0-3). The Cyclones were trying to bounce back after a heartbreaking last-second 45-42 loss to No. 13 Baylor last Saturday and get the first conference victory for first-year coach Matt Campbell.I thought it was a gutsy performance, but we just have to learn to finish, and thats part of what were doing, Campbell said.David Montgomerys 1-yard touchdown run with 2:58 left in the second quarter put Iowa State up 17-14. The Cyclones nearly doubled time of possession in the first half and ran 55 plays to Oklahoma States 32.Joel Lanning threw touchdown passes to Trever Ryen and Allen Lazard to help Iowa State take a 17-point lead in the third quarter.I warned my team last week about Iowa State and the ability they had to move the football, Gundy said. (Their) quarterbacks are good players and they spread you out and are good runners. Theyve got good schemes on both sides of the ball, and they made some real nice adjustments on us.Rudolph threw a 35-yard touchdown pass to Washington, then a 1-yard touchdown scoring toss to Jalen McCleskey, to help the Cowboys trim Iowa States lead to 31-28 early in the fourth quarter. Oklahoma States Ben Grogan made a 30-yard field goal to tie the score at 31 with 11:33 to play.THE TAKEAWAYIOWA STATE: The Cyclones proved once again that they are competitive, but arent quite there yet. They got solid quarterback play at times from Lanning and Jacob Park, but neither was consistent, and they were sacked a combined seven times.We have the capability and skill set, its more of a process of learning how to execute and learning how to win, Iowa State cornerback Brian Peavy said.OKLAHOMA STATE: The Cowboys built on their victory over Texas the previous week and enter their off week with momentum.Its good to see your guys keep fighting, Rudolph said. I dont know how many points we were down, but this place was kind of dead. The defense played a great game in the second half and got us some stops and turnovers that were much needed. We didnt turn the ball over, which was big for us. We just kept fighting.POLL IMPLICATIONSOklahoma State was 32nd in votes in the most recent poll and likely will be knocking on the door of the Top 25.UP NEXTIOWA STATE: The Cyclones play at Texas next Saturday. The Longhorns have has allowed 50, 49 and 45 points in their past three games. Iowa State could deal a severe blow to struggling Texas coach Charlie Strong with a win.OKLAHOMA STATE: The Cowboys are off next week. They will play at Kansas on Oct. 22. Gundy said it is a good time for a bye week because several of his players are dinged up.---Follow Cliff Brunt on Twitter (at)CliffBruntAP . Wholesale Adidas Ireland . Hargreaves began his career in 2008 with the Winnipeg Blue Bombers and has played with the Edmonton Eskimos and last season with the Saskatchewan Roughriders. Clearance Yeezy 350 Ireland . Dukurs winning time was 1 minute, 45.76 seconds, a quarter-second better than Russias Alexander Tretiakov. Lativas Tomass Dukurs was third, 1.41 seconds off the pace. Jon Montgomery of Eckville, Alta. http://www.yeezy350saleireland.com/ . -- The goal posts lying flat on the field, Arizonas fans lingered on the field, congregating around the locker room entrance nearly 30 minutes after rushing out of the stands. Cheap Yeezy 350 Ireland For Sale . Terms of the deal were not immediately available. The 26-year-old finished last season with Clevelands Triple-A affiliate in Columbus after signing with the Indians in August. Wholesale Yeezy 350 Ireland . Neymar curled home a free kick from just outside the area to put the 2014 World Cup host ahead in the 44th minute. Three minutes after the break, a simple through pass from Paulinho freed Oscar and the Chelsea star rounded goalkeeper Jung Sung-ryong to extend Brazils lead. One debate ended in St. Louis on Sunday night, while another debate shows no signs of slowing down in Texas.The 4-1 Cowboys are one of the most surprising teams in football, and at the center of their success is rookie fourth-round pick Dak Prescott, who has surpassed even the lofty expectations augured by his promising preseason. Given how previous preseason sensations looked when the gloves came off, I was skeptical that Prescott would be able to live up to his level of play from August. I was wrong.Adam Schefter reported Sunday morning that the Bengals, Prescotts opponent in Week 5, broke the rookie down on tape and did not believe he had thrown?a single bad pass all season. I would argue that there is probably a pass or two in there which might fit the description, but thats beside the point: Prescott has been incredibly effective. Outside of some aggressive decisions in the red zone, Prescott also didnt do much on Sunday to prove the Bengals wrong. He was phenomenal during Dallas 28-14 win, going 18-of-24 for 227 yards with a passing touchdown and a rushing score on a read-option keeper. Prescott fumbled in the red zone and took an unnecessary shot from Vontaze Burfict late in the game when he neglected to slide, but he finished with an 87.5 QBR, the fifth-best figure of the week.Prescott is now second in the league in opponent-adjusted QBR at 84.1, behind league leader Matt Ryan, and he might be losing his job in a few weeks. Even after this hot start, owner Jerry Jones has insisted that Tony Romo is still the teams No. 1 quarterback. Romo is recovering from a broken back, and while the Cowboys obviously dont need to rush him back into the lineup, the preseason timeframe for Romo suggests that hell be back after Dallas Week 7 bye for what now looms as a crucial divisional matchup against the Eagles on Oct. 30.Should the Cowboys insert Romo back into the lineup? Or should they keep Prescott -- who is eight passes away from breaking?Tom Bradys record for most pass attempts to start a career without an interception -- in the job for the foreseeable future? I think you can make a case for each side, so lets do that and see if one seems stronger than the other. Starting with the incumbent:The case for starting Tony RomoRomo has been a better passer than Prescott. As good as Prescott has been this year, when Romo has been healthy over the past couple of seasons, he has actually been a tiny bit better. Heres a table comparing their numbers since the beginning of 2014, when Romo was gifted with Zack Martin to complete Dallas devastating offensive line:Prescotts QBR is better because he has added significant value as a runner, which counts, but might not be sustainable over the long haul. Prescott has three rushing touchdowns in five games, for one. He also wont manage to go his entire career without throwing an interception, and even while we give him credit for having done so up to this point, Prescotts passing numbers are slightly worse than Romos?on a rate basis. The biggest difference is that Romo throws downfield far more frequently than Prescott, as noted in the air yards column; 11.2 percent of Romos passes traveled 20 or more yards in the air, compared to just 5.2 percent of Prescotts passes.There are pretty huge error bars on both sides of those numbers: Romo is 36 and missed most of the past year with injuries, while Prescott is five games into his NFL career. Given what weve seen most recently, though, Romo has been the slightly better passer. Given how Romo spreads the field more frequently, he also seems to be the quarterback with the higher short-term upside for the Cowboys.Romo is more experienced. If the Cowboys are trying to win now and value experience as an asset, Romo obviously has that over Prescott in spades. The old arguments that Romo would turn to dust in December were always foolish and mysteriously never came up in 2014, when the Cowboys went 4-0 and Romo threw 12 touchdowns against one interception with a 133.7 passer rating. Romos playoff record is an underwhelming 2-4, but he has been relatively effective within those games, throwing eight touchdowns against two picks and posting a passer rating of 93.0. His sack rate in those games is a staggering 10.6 percent, nearly double his career rate of 5.4 percent, which might tell you what has gone wrong in those contests.The argument against rushing Prescott into a key postseason role before hes ready for the moment, ironically, is Tony Romo. It was Romo who had critics bestowing him with the it factor during the red-hot start to his own career before a pair of brutal postseason blows. Romo famously fumbled the snap on a chip-shot field goal attempt which would have sealed a playoff win over the Seahawks during his debut season in 2006, and he lost to the Super Bowl-bound Giants at home the following year, which cemented his unfair reputation as a passer who couldnt win when it really mattered.Prescott may very well win those games when he gets the opportunity, or he might fail if the Cowboys give him the shot two or three years down the line. It also helps that Prescott looks about as cool and comfortable on the field as any young quarterback in recent memory. If experience matters, though, Romo obviously has that point in his favor.Romo is probably sticking around for 2017 either way. I wrote about Romos contract situation before the season, and while the Cowboys have locked up center Travis Frederick to a long-term deal, not much has changed. Dallas already has nearly $173 million in salaries tied up on its 2017 cap, and thats without considering deals for Barry Church, J.J. Wilcox, Brandon Carr, Terrance Williams?or even Morris Claiborne, who has shown flashes of competence this year. (Even if the Cowboys dont want to re-sign any of those guys, they still need somebody to start in the secondary.) Dallas will create cap room by restructuring Tyron Smiths deal for another year, but theyre still going to be working with limited salary-cap space for yet another season.After restructuring Romos deal twice since re-signing their star quarterback in March 2013, the Cowboys couldnt make a clean break even if they were so inclined. His cap hold on the roster next year is $24.7 million. If the Cowboys cut Romo, they could designate him as a post-June 1 release and eat $12.7 million in dead money in 2017 with $6.9 million in additional funds hitting their 2018 cap. More plausibly, if the Cowboys trade Romo or encourage the oft-injured passer to retire, Dallas would be responsible for $19.6 million in dead money for Romo on its 2017 cap.Could the Cowboys afford that sort of cap hold? Its not out of the question. The Cowboys have Prescott locked up for the next three seasons after this one with cap hits between $635,848 and $815,849. If Dallas doesnt need to find a starting quarterback, it could absorb the massive hit from a Romo trade next year, use the $5.1 million in savings to sign a backup quarterback to replace Romo, and then enjoy all the benefits of a bargain-basement quarterback in 2018 before locking Prescott up to an extension in 2019.Given that Romo will be 37 and coming off two injury-marred seasons, though, will there be an enormous trade market for his services (and contract)? It wont help that the Cowboys will have little leverage, since teams will know Dallas will be trying to offload its star passer. Itll also be in a market where organizations could be considering veteran options such as?Sam Bradford, Jay Cutler?and Colin Kaepernick in trade or as cap casualties. Somebody will be interested in Romo, but if all the Cowboys can get is a midround pick (a la the Brett Favre trade from Green Bay to the Jets), wouldnt it be better to keep Romo on the roster and have two viable quarterbacks on hand at all times?The case for starting Dak PrescottPrescott hasnt done anything to lose his job. Theres no simple, concrete answer to the question of whether a player such as Romo should lose his job if he gets hurt and the replacement does well, as you can see from this article. Some players who have been on the injured side of that situation and fearful for their futures would surely tell you they deserve a crack at the gig once theyre healthy, while athletes who lurked on the bench and desperately waited for their opportunity would likely say the opposite. Its a matter of preference and context.With that being said, theres a certain level of credibility a coach gains -- with fans and inside the locker room -- by self-evaluating and putting his best players on the field. Jasson Garretts job isnt in danger, but he has pulled all the right strings during Dallas 4-1 start.dddddddddddd If you watched the Cowboys blow out the Bengals on Sunday, you saw an unfamiliar scene among the guys wearing blue and white: smiles all around. Even the injured players on the bench like Romo and Dez Bryant were beaming. Im not saying the Cowboys will lose those smiles or see their chemistry dissipate if they make the move back to Romo, but if they do, and it turns out Prescott was the better option, it will cost Garrett a serious chunk of credibility.Romos margin for error will be razor-thin. I mention this a lot when we talk about teams promoting high-profile rookie quarterbacks into the starting role, but it applies here. The Cowboys dont need the fans on their side to win, but if Prescott plays at this level for half a season and is dropped to the bench for Romo regardless, its going to dramatically up the tension at Cowboys games. A not-insignificant percentage of the Dallas fan base already holds some level of contempt for Romo. Everyone loves Prescott.Lets say Prescott continues at this level of play for three more games and then the Cowboys go back to Romo. The first time Romo screws up -- he misses an open receiver, he throws an interception, even pieces together a few consecutive scoreless drives -- what do you think will be the first thing Dallas fans chant? We Want Dak, of course. Very few fans would insist on Romo if Prescott plays at this level and suddenly has a bad start in Week 8. The one thing the Cowboys cant afford to do is go back and forth and erode each quarterbacks confidence. Once Garrett makes a switch, he has to stick with it for a while. That alone may be enough of an argument to keep Prescott in the lineup until he has proved he cant handle the job.Prescott is Dallas quarterback of the future, and once you find that guy, you build everything around him. Even if Romo might narrowly be the best option for the next season-plus -- and that is very debatable -- Prescotts clearly the guy whom the Cowboys are about to build their next 10 seasons around. He may not end up rising to that level, but Prescott is the first passer to come around since Romo whom the Cowboys view as a franchise-caliber asset.Given how important quarterbacks are, a properly incentivized Cowboys team might find its most important short-term tasks after winning a Super Bowl will be figuring out whether Prescotts the guy and giving him as many reps as possible to improve. Obviously, Prescott is far more likely to learn and improve in the starting lineup than he would be watching Romo while holding a clipboard on the sidelines.Leaving Romo on the bench might actually increase his trade value. If the Cowboys have decided that Prescotts their guy and they dont want to threaten him with Romos presence on the roster, the smartest thing to do is trade Romo for a draft pick and clear up his dead cap holds as much as possible to free up space for 2018. And given that Romo may not be the same player he was even two years ago, before these injuries, the Cowboys might prefer that other teams most recent professional tape on Romo comes from the 2014 and 15 seasons as opposed to the games featuring the man with the fractured back. If Romo has slipped and the Cowboys see it in physical therapy or practice, the presence of Prescott allows Dallas to retain plausible deniability if it wants to deal Romo. Being able to hide Romo from view may be the difference between getting a fourth-round pick as opposed to a conditional seventh-round pick. Theres value in that.I can see bits and pieces of both sides here. Each side has a case. The more I think about it, in fact, a third possibility came to mind. (No, its not Mark Sanchez.) It might be the least satisfying answer of all, but its also the most accurate one from my perspective:The case for it not really mattering who the Cowboys start at quarterbackTheyre both good. While there are still question marks around Romos health and Prescotts relative lack of professional experience, everything we know suggests that Romo and Prescott will continue to play at a high level when in the Dallas lineup. The table I included in the Romo section says that he might be the better passer, but the numbers are close enough and at a high enough level that theres not really a wrong choice between the two. Its not as if the Cowboys are picking between a hot young quarterback and a league-average veteran, as the 49ers did when they chose Kaepernick over Alex Smith after the latter missed a couple of games with a concussion and Kaepernick excelled in his absence. These are two quarterbacks who can play in reasonably similar offensive styles and produce passer ratings in excess of 100.This team is built around the running game regardless of its quarterback. Dallas run-first identity isnt some new game plan the team installed to make things easier for Prescott. Ever since using high picks on Frederick and Martin, the Cowboys have gone with a run-heavy offense and built their team around that dominant rushing attack. During Dallas 8-8 seasons from 2011 to 2013, Romo averaged 36.3 passes per start. In 2014, when the Cowboys were defending more leads late and turning the ball over to superstar DeMarco Murray, Romo averaged 29 pass attempts per game.This year, with the Cowboys giving the ball to the phenomenal Ezekiel Elliott, Prescott has averaged only 31 throws per start. After a slow beginning to his professional career, Elliott has kicked into another gear. He has now produced three consecutive games with 130 yards or more, and unlike Week 3, when the Cowboys needlessly ran him into the ground for 30 carries in a blowout victory over the Bears, Dallas gave Elliott only 15 carries on Sunday before letting Alfred Morris carry the rest of the load. Im still not sure Elliott was worth the fourth overall pick given how good this line looks -- you can imagine the Cowboys might be very happy if they had chosen Jalen Ramsey with the fourth overall pick and spent a fourth-round pick on Jordan Howard -- but with the draft capital a sunk cost, Elliott looks great.If the defense doesnt hold up, the quarterback wont matter. This Cowboys team is capable of winning a shootout with just about anybody when its main assets are healthy, although its perpetually trying to keep those stars active. Just as Tyron Smith was gone for a couple of weeks before returning to the Cowboys this week, Dallas has lost Bryant to a knee injury which could hold the Pro-Bowl wide receiver out until the bye week just like Romo.What has really changed for the Cowboys this year, in addition to improved quarterback play over the Matt Cassel-driven disaster of 2015, is the presence of a much-improved defense. The much-maligned Cowboys pass rush doesnt have any stars, but Dallas is still managing to pressure opposing quarterbacks 23.6 percent of the time, which is a respectable 22nd in the league. Even more notable is that the Cowboys are finally creating takeaways again. In 2014, the Cowboys ended 16.8 percent of opposing possessions with a takeaway, the highest rate in the league. In 2015, that figure fell to 5.6 percent, the lowest in the league. This year, their takeaway rate is 10.4 percent, which is a respectable 18th.If this team can run the football and play average defense, it can begin to approximate a winning football organization without even being concerned about picking a passer. And if the Cowboys cant run the ball or play defense, Prescott and Romo will each end up in a lot of third-and-long situations with no help at all, which doesnt play to their strengths whatsoever. This speaks to the value of building an infrastructure around a quarterback to make everything else work. Passers such as Prescott and Carson Wentz are off to remarkably quick starts, and the effectiveness of their offenses has much to do with quality offensive lines and well-schemed game plans. Jared Goff and Andrew Luck would kill for that sort of help.Knowing that the Cowboys will need to make a choice one way or another, Im inclined to lean toward keeping Prescott at quarterback for now, if only because its a lot easier to go from Dak to Romo if the former struggles than it would be to go from Dak to Romo and then back to Dak a few weeks later. As I suggested in writing about Jimmy Garoppolo a few weeks ago, though, its a problem the Cowboys are blessed to face. ' ' '